Since finding a way how to report an issue is something that's been made extremely difficult by the reMarkable on their website, I'm sharing a link here where apparently (as far as I can figure it out) we can fill out the form and send the ticket: support.remarkable.com/hc/en-us/requests/new?ticket_form_id=360000022918
Thanks for checking this issue. It happens on mine as well, just reported it to support (although I'm not hopeful they will address it for months to come or that they will give any kind of timeline, given all past experiences).
Thanks. Just received mine (batch 3) an it also has the issue. Just reported it to support and asked for advice on what should I do either ask for a replacement or wait for a software update.
Ohhh, this is great, thanks! The more people we can get bugging them about stuff, the better our chances. Videos online from guys like you are even a bigger motivator to them, so thanks again for those too. Keep up the great work! 👍
Hi Voya thanks for the effort and honesty as always. I believe the question for me (and some others) is whether the device is still a better choice when compared to others (Boox air for example), especially for the ones who are waiting, since the selling point until now has been the writing experience - and if the writing is jagged I feel like the writing exp overall would not be that wonderful anymore, and together with the wait, the lack of some functionalities and tools - does it still give a run for its money?
@@NecroDomoEPI I have just sent the issue to them as well, along with the files I have exported and showed in the video, an explanation of the issue, a link to this video, and the report from the Facebook poll that out of 97 who responded 69 are affected by the issue, which is 71%. This is far worse than I initially thought because I thought that it would be the other way around. So, we'll see if they respond and address this in any meaningful way. I hope that they do because this affects directly their main selling point and the main advantage of the device in relation to the competition. So it would be very unwise to ignore it, and not treat it with high urgency and importance.
The best thing we can all do is report the issue so many tickets turn into a software (post line smoothing) fix ... excellent uber detailed analysis as ever. Thanks!
Voya (I hope I got the spelling correct), you are unique in our field and a true public asset. Your technical knowhow is exemplary. However, your best asset is your advocacy persona. I work in medical research and often connect with patient advocates. Finding that balance of technical knowledge, clarity on how an issue affects different groups differentially, and passionate advocacy for the group as a whole is what makes and advocate a true public service. You have convinced me to support you and your work through the patron program as you are a vital public service. Thank you!
I'm not returning my Remarkable 2 for sure as it is very usable, but I agree with you that Remarkable must step up their software game. But this might be so simple as settings tweaks or it needs some filtering.
Or just a calibration, there are many possible fixes, I just hope they do it soon, coz now I can't use rM2 for one of the main purposes I needed it for :/
Voja you are amazing. All that scientific analysis AND a philosophy lesson as well. It's strange how hiding behind a keyboard gives riteous indignant types the license to 'blow off' a company over one bug. Your 'stay calm and just report the facts' was EXCELLENT advice in this and many other situations. Well done and also for saying that you Love the device (RM-2).
I believe reMarkable would have been well aware of this behaviour, how could they not. Fixing the issue was either going to make honouring delivery dates impossible, or would negatively affect the user experience so badly that they decided to go ahead with a compromised product and worry about it later..., maybe both! I'm concerned that they won't see it as something that needs to be fixed at all, but just how their tablets work in order to achieve the low latency and paper like experience. I hope reMarkable responds quickly and positively next week, otherwise I might just stick to real pens and ink for a while longer. Many thanks again Voja. I much appreciate your honest and comprehensive reviews.
Batch 8 here. Still haven't gotten it. But please, let us know any updates on the issue (answers, timelines, etc.). I've grown worried about this and about the reading capabilities of the device... I'll wait for mine and probably fill in the ticket as soon as i get it. Thanks for your work!
@Voja I had an idea directly, but I'am not so experienced in *svg. Do you think the huge amount of point's and resulting splines are responsible maybe for the low max. page storage(7000 instead of claimed 100000). May you check how big the svg from 10 horizontal lines are on the rM2 vs. rM1(If not I understand and no comments needed). The lag of storage was one reason why I cancelled my pre-order. There are some insane compression algorithms on the market but maybe they are not using them or there own isn't good enough. Great video and very interesting. In some cases you are really deep investigating, 10 times more than me. I'am learning a lot from you, and even if I would/should do it the same, I often enough don't do it. It's really nice to watch your work and reflect about it. Thank you for your efforts!
I also have Remarkable 1 and 2. I have observed very mild horizontal jitter lines on R2 which I am able to virtually eliminate or make worse by very simple methods. Please note, I am not suggesting this will work for everyone, rather it may give some clues to those with a better grasp of the physics of these pen/screen systems regarding what may be going on. YMMV. As soon as I saw your posting I was reminded of the experiments I carried out on R1 with transparent (physical) drawing templates. I was interested in whether I could cut designs, or geometric shapes, through thin transparent plastic sheets (like tech drawing templates) to speed up some drawing tasks. I quickly realised the pen body could easily ride up on edges of cut out templates, curves or rules, unweighing the nib, which stopped drawing. I therefore moved on to clear sheets of Lexan that were not engraved right through but would guide the nib. I observed there seemed to be an optimum thickness of template sheet for getting smooth lines. I found that as the height of the pen above the surface rose too high, the lines could become progressively more jagged, until they look like seismograph traces. It is nonetheless interesting to draw on clear plastic a few mm thick and watch the lines appear well below the pen tip. Incidentally, I have previously tried various other template materials and found interesting effects .Metal edges for example could displace lines and slots just wider than the nib in thin metal shim gave (non displaced) jaggies even when the nib was on the R1 surface. Anyway, today I found using my R1 pen to rule lines on my R2 the jaggies (which are not very bad to start with) almost disappeared. On inspection, I noticed the R1 pen nib was warn down, by a measured 0.21 mm. I swapped the nibs between pens and confirmed the short nib eliminated the jaggies on both pens when used on R2. I then got out the plastic sheets I had used for my previous experiments and was able to induce proportionally worsening jaggies on both R1 and R2 , with both pens/nibs, in any direction, by increasing the pen height above the surface. I did observe the reported relative horizontal/vertical line differences between the R1 and R2. My observations suggest that depending on the angle of the pen, the height of the pen body above the surface can be quite critical, in relation to how many points get recorded while ruling a line on an axis. Small variations in manufacturing tolerances may be summing or subtracting to account for whether the jaggies become annoying or not, in addition to whatever filtering or smoothing may be going on. I will be interested to hear if others have any luck with worn nibs of varying lengths.
Very interesting And thank you for sharing :) it could be a thing, that the screen protector installed on top is inconsistent in a way. However, have you tested the pressure, because for me, it was the main thing that made it worse (regardless of the pressure or tilt sensitivity of the brush) pure physical pressure results in worse jaggies, for me and quite a few others as well.
@@MyDeepGuide Yes, I also found that increasing pressure gave more/bigger jaggies with the pens I have tried: fine liner, ballpoint, tech pencil and calligraphy nib. I tend to use the fine line pen for most writing and diagrams, having previously enjoyed using Rotring tech drawing pens and /or 4B thin mechanical drawing pencils on drawing film. Since posting above, I have experimented further. I progressively filed down a nib from 2.88 mm (shoulder to tip) to 1.8 mm but did not see any marked improvement over the 0.2mm wear I started with. I think the R1 pen may be marginally less prone to jaggies but the difference is very small and only evident with the shorter nib. I got the best looking horizontal lines drawing on R2 surface using the fine ballpoint nib, the 0.2 mm worn nib and a light touch. The magnitude of jitter shown was less than a quarter of line thickness. I observed variation between one line and the next, however, which got me thinking about sources of noise and resonance and how they may interact with the various analogue to digital to digital conversions. I am also involved in photography and have encountered interesting issues converting jpeg (raster images) into vector images, where matching sampling frequencies and adjusting filters can be crucial. To explore some likely sources of noise leading to jitter. I used the fine liner on R2 with R2 pen and a new nib, but I placed a variety of films on top of the R2 surface, intending to modify surface induced vibrations. I also varied pen pressure. The results proved very interesting. 0.12mm, very smooth mylar sheet, with light pressure made almost perfectly smooth lines. Increasing the pressure produced some intermittent "stiction" with occasional low-medium amplitude jiggles, depending on pressure, along with some low amplitude higher frequency noise. The paper drawing film stuff (0.08 mm) the Remarkable came wrapped in, produced very uniform, meduim to high frequency, hi amplitude jaggies which varied less with pressure. Textured 0.54 mm plastic sheet that has a similar to, but scaled up surface to the remarkable, produced similar results to the remarkable but with even lower amplitude, low spatial frequency, jaggies on light pressure and even higher amplitude on higher pressure. As a result of these brief tests, I am inclined to think that the pen nib is picking up surface texture (including the intentional "paper like feel" vibrations on R screen surface) and/or frictional vibration. Maybe due to mechanical resonance and/or sampling effects, these vibrations are being interpreted as horizontal movement, maybe due to the apparently high resolution of the sensor, to show up as jitter. Presumably there is some crucial difference in how vertical and horizontal movement is processed. These ideas may explain why varying the pressure on the nib has such a big effect on level of jitter. Like the preload on a record player pickup cartridge. So far, the magnitude of any problem is minimal for my everyday use of Remarkable. I usually draw with a light touch in any case. If I need to draw precise, straight horizontal lines, I can always lay down some Mylar sheet. Indeed, my past experiments on R1 led me to make a thin Lexan ruler, with a smooth V groove engraved, almost all the way through, which constrains the pen nib and facilitates smooth straight lines. I was able to do some very precise drawing by using the powerful 500% Zoom facility, so long as I stayed away from the edges where things get non linear. I explored using the remarkable as a tech drawing input device by placing it on a mechanical drawing board. The idea was to convert the R1 input to vector then clean it up if needed on a computer. The intention being to use familiar, fast, tech drawing methods and physical tools to bridge the gap between a rough sketch and CAD. I explored making some mini mech drawing tools sized to the R 1, using linear digital scales and digital angle measuring tools. So far it still looks feasible, assuming the necessary desktop software conversion tools can be made to work reliably. I am keen to see what you and others may find there. Like you, I think, hope, there may be some tweaks to the sampling/rendering process that might be useful in time. So far I don't have the time to explore a vector export workflow in depth. Similar issues may also occur in photography. There have been debates over anti aliasing filters on sensors vs post processing solutions and so far we seem to have moved in the direction of the latter. Again I remain unsure whether my findings will generalise to others use cases. Keep up the great work!
Your channel truly lives up to its name. You've enabled the pickle to be both deeply ruminated by myself and pickled in the rumens of collective ruminants in the same stationary ecosystem. Let us anticipate the cultivation of patches with less messy pickles for the future.
Good analysis. Sensible approach to recognizing bugs do happen. I agree that this is an opportunity for the company to demonstrate their responsiveness. They have been extremely generous with people exchanging defective devices for good ones, so I believe they are capable of matching this on the software side. They do need to prove it to keep our goodwill, however.
Thanks for putting this on my radar, Voja! Otherwise I could've overlooked the issue in the trial period as I have finally received my device today, and it's my first e ink device. It's a truly beautiful piece of hardware but unfortunately I have the issue as well, although at first I thought I didn't. Two take aways: 1. It does indeed get worse if the pen is perpendicular to the screen. 2. Pressure adds to it. 3. It gets truly unworkable for me on the bottom cm of the screen, add to that that the line seems to be offset by 1 or 2mm in that same area. I've sent a support ticket but I truly hope it can be fixed by calibration or a software fix, otherwise I'm almost certainly going to send back or sell the device. Probably the first as I don't want to resell a defective device. It's way to expensive for that. It for a large part takes away the joy and excitement I had opening the package as it's uncertain if I can even use this in my daily routine. Edit: batch 5
I wonder if this is the result of optimization for speed (lowering the latency in response to the pen) on the rM2 at the risk/sacrifice for smoothness in tracking the pen. That is to say, they may have changed the interpolation of points when a line segment is drawn from point A to B. Also, just also wanted to add that given their build development environment you can still have the same OS version available for both devices, but you can have MACROS defined that would take advantage of the target hardware given that the rM2 has a slightly more "powerful" processor and double the memory.
Probably this is due to the fast screen reaction/responsivenes of the RM2. They would have to "fix" the line some seconds after you drawing, as filtering needs time and data from the "future" points. If doing so, you get the impression, that the line changes after drawing. That could be not so "paper like". So it is somehow a trade of. But for sure, that needs a lot of work software wise. And: Thank you for your deep dive into that!
what kind of company allows product to go out their door like this ? I've been waiting 6+ months now for the rm2 and this excellent video just made me lose my cool. Sheesh! I feel like I've been robbed by an eloquent pick-pocket.
@@MyDeepGuide if those that don't have it REALLY don't have it then that points to a hardware issue. Maybe the installation of the screen is faulty ? I've seen reports of the surface of the screen protruding unevenly above the surface of the casing.
I'd like Remarkable teams start to improve his comunication. It' ok have some issues, but it's their job not only fix issues, but even advice us about their plan for the next months. Because if they don't want to share our future I'll send back my remarkable 4 when it arrives (batch 4).
Voya, i'm very impress with the quality of your analyse. I'm waiting for my remarkable 2 (batch 4) and your video help me to wait. Hope that they resolved this problem soon.
Thank you for watching :) Their response they sent out to users reporting on the issue (i haven't received a reply though for now 10 days, i guess they are no longer talking to me even to the support tickets), isn't the most encouraging one, as they basically downplay the issue and say that it doesn't "impact most users", even though it seems that the majority of the devices (or all) are affected. So yeah, I'm not sure they are taking it seriously, based on their response.
Thanks for this clear and deep analysis of the situation. It is understandable for a person like me who is not a technician 😂 I really appreciated the advice you give at the end telling people to report gently the problem, I do believe that there's no need to be angry in order to get satisfied 🙂 I'll keep my RM2 recieved yesterday since I'll use it essentially for writing, and I'll report the issue 😉
@@MyDeepGuide But if this is really the case, wouldn't it make more sense to try to get a replacement? I mean ideally we would love to have better raw data than smoothend post-processed curves....
My guess is that the screen refreshes and scans for inputs in a certain direction at certain intervalls. Due to the different form factors of the reMarkable 1 and 2 screens, those directions might be different (vertical on the R1 and horizontal on the R2). The reason the issue is present on the R1, but worse on the R2 might be caused by the higher "refresh rate", as in: lower pen latency of the R2. So I reckon it might be the exact same root cause on reMarkable 1 and 2 and is actually a software issue that can be fixed via an update.
Absolutely a very likely culprit. It's also possible that 4096 levels of pressure sensitivity instead of 2048 come into play as well, because data is being piled up almost one on top of each other. Or a combination of both. Either way, post processing optimization filter, especially on a dual core is a background process that shouldn't affect the performance of the device, yet would offer immensely better end results, not just for SVG, but for PDF and PNG exporting.
@@MyDeepGuide While post-processing would just be a band-aid for this issue, I agree that it's a good place to start as it addresses so many other issues, like file size and thus large notebook performance.
One would think that this kind of a conversation was supposed to be happening in Remarkable's R&D offices this time last year while they were developing the rM2, and not on a TH-cam channel, 2 months into shipping the units to users. Strange...
@@MyDeepGuide then wouldn't it be interesting to know if users that do NOT have the issue on their RM2 may just try vertical lines instead? Because _maybe_ the panel or digitizer manufacturing can be oriented both ways? Also another thought: What if this was an intentional way of reducing the latency? I mean: more points per second would mean lower latency. But also more points, as we see now, also translate to fuzzy curves
@@faroit I'd bet money that people who don't experience this effect on their R2 have it with vertical lines instead. It's probably not intentional, but the lower input latency amplifies the effect.
I was really looking into buying this, but after this I am done. Thank you! I am pissed off that they lied about Remarkable 1's marker not working on the 2nd generation. You can find it on their site. Cool idea, really wanted a writing only tablet, but... too bad.
@@beyondbased I understand that and I agree. The pressure sensitivity is important, but look at this: "Does the Marker for reMarkable 1 work with reMarkable 2? No, the Marker and Marker Signature for reMarkable 1 are not compatible with reMarkable 2. The new Markers are calibrated to work with reMarkable 2’s second-generation CANVAS display" It s a marketing scheme. The answer is valid, but the question is there to induce confusion.
I found this commwnt from facebook group. I can tell this guys knows what he is doing :) Quat: "A long time ago, I worked on a high end commercial illustration program called Aldus FreeHand. It was the head to head competitor with Adobe Illustrator at the time until Adobe started gobbling up everyone. Part of that work included creating bezier curves to a set of points and also create outlines from bitmap images. When I look at the jagged lines problem that shows up on RM2 but not RM1, I see that the RM2 is capturing points at a much higher rate than the RM1. These points also seem to have a higher "variation" of being to the left or right of the pen path than the RM1 did. It makes sense to me that there are more points captured because the RM2 is running faster than the RM1. The work I did converting bitmap images to lines tells me the jagged lines on the RM2 can be cleaned up with a software optimization. You can fit a straight line to that mess of points collected and you can also selectively eliminate some of the points collected. What I don't know is how whether the hardware gives consistent results from one pen stroke to the next. If the results are farily consistent for any pen stroke (i.e. you get the same number of points and they create the same jaggy points), then it's a straightforward software fix. If the hardware is acting more randomly, you might still be able to fix the issue easily or it may require a bit more work. Another fix is that RM can probably tune the rate at which it collects points for any given pen stroke. This will reduced the fidelity back toward RM1, but might be an easy quick solution."
It's a bit more complicated than that, and the best software solution would be 3 steps: 1. Gentle low pass filtering to take the wild peaks and put the jittering into a more equalized noise. 2. Point proximity optimization, which would determine the distance between points (any with any in a given stroke, not just consecutive ones) and weld together the overlapping points or those that are within a tiny and carefully calibrated threshold radius. 3. Finally, the most important step, sampling the optimuzed stroke and rebuilding it using bezier curves to get the closest possible match (can be pretty much perfect). This is basically handing out the solution to the reMarkable team how to fix this on the secondary thread or as a regular coroutine as a post process after each stroke is completed. But they won't bother with it, that much is clear from their response.
Shoot! One of the main reasons I have the ReMarkable2 on order, and that is for drawing. I am so sick of sketch books everywhere. I guess I'll have to see how bad it is when I get it. Thanks so much for the deep guide into this issue.
Great video, as always. Got my Remarkable 2 today, hopefully, mine does work properly, if not, hopefully, they will fix it. Thank you for all your content!!!
Very well done, as usual! Such a pragmatic approach. I must confess, I can get a bit firey on the Facebook pages, and I think that's amplified by how much I 'want' to love this thing. I mean the hardware is just... Stunning! I'm also keeping my preorder, and hoping to god that this is fixed soon 😁 I've said it before, but the hardware design team are basically saving this company right now!
I’m looking into purchasing the Remarkable 2, but after watching this video I’m starting to have a second thought. You said that Remarkable 1 and 2 use the same software, but then you said that we don’t know if it’s a software issue or a hardware issue. However, if it’s not having much of an issue in Remarkable 1, then is it a hardware issue? Also, this great study you’ve made was made a year ago, and I wonder if the issue been solved. Thanks, Jake
Hello, This issue has since been resolved, because if you look at the video date, you'll see that it's quite old. But this is no longer a problem on the Remarkable 2.
I particularly appreciate the tutorial on how to behave when commenting or reporting issues. Thank you Voja. (And... Nerdy things! 😊 The vector format check)
Thanks for posting this video showing in detail the problem. The issue is easy to replicate and so I have logged a ticket with remarkable too. I really hope this is fixed because it seems the pen feel of this device is unmatched by their competitors (even if the functionality is lacking in comparison).
Remarkable are pariahs. I'll test it out but there bs answer -no answer really took several weeks to get a response that was not a response. Love your content. Any thoughts on amount of magnets in device that could cause issues for mechanical watch?
Please guys fill the ticket, if we are only few of us they ll ignore it for sure. Btw thank you for your review I'm always impressed by how you do it well.
Thanks for the detailed look into the problem. It could be related to the faster response rate on the RM2. If there's less time to sample the location of the marker then the location accuracy will be lower. If that's the case it could easily be fixed in software by applying a little bit of filtering, though that may make writing small text more 'blunt'.
I have the issue, but as I use the RM2 mainly for notetaking it doesn't bother me. But I did notice trying it out is that there is a difference in the amount of jaggedness when you tilt the marker. When you tilt the marker while drawing a line the jaggedness is substantilly less.
I'm having a similar issue with my Max 3 (although not that bad). I've been using my Max 3 for the past 2 months and love the device. I notice my traces are a bit shakey though. Someonw said it has to do with the inaccuracy of the EMR screen. Is there a setting I missed that can smooth out my traces when I write?
Second day of ownership for me. I tried and confirmed the jaggedness. It can be smoothed out to a degree by creating the lines in zoomed view. It is a workaround but does not require extra software
An unrelated question: how long does it take to fully charge RM2(e.g. battery from 8% to 100%)? It took me 8 hrs to get from 8% to 85%(not 100%). I am not sure if it is only me or it is usual that RM2 takes long time to be fully charged.
I’m 17 minutes into the video. I have to leave for work, but I wanted to share some initial thoughts: 1) Just because you have the same OS version between the rM1 & rM2, it does not mean that the two devices share the exact same digitizer hardware, digitizer firmware, and/or digitizer drivers. I’ve seen this kind of issue in early android phones. The nexus one for example, could not draw straight lines very well at all, and the lines were very wavy. This kind of issue could stem from badly designed or faulty digitizer hardware, or a bad digitizer firmware, or a bad digitizer driver. It’s also possible that different rM2 Devices are being shipped out with different digitizer hardware from different vendors.
22 minutes into it now. A couple of additional thoughts: * You tried varying pressure, but not angle. * Some people in the comments are confusing antialiasing (which I think the Boox devices do) with point smoothing (not a graphical process change, but just a change to the way the point data is interpreted). I wanted to clear that up, but I didn’t have time.
21:46 dang, that’s some crunchy data. It’s as if it was filtered through a guitar distortion amp 😂 Clearly, some corners were cut, either in the hardware, firmware, drivers, or software.
Yeah, this is beyond wavy lines and AA. It looks like the raw input is very garbled, especially the triangle, and yeah it can be anything at this point. Post processing proximity filtering, optimization and smoothing can be a good temporary patch, helper thing until they figure out what's really going on, but evidently, some unites have this, some don't. Would be good to get a rough ratio estimate to understand better what's up.
I tried angle, but video was already long enough, no difference. That's also why I used the Fineliner, coz it's the bread and butter brush, no pressure, no angle.
I am pretty sure it was already mentioned, but this "jaggy" lines look like an issue between the higher sensitivity, the different surface and the capacity of the pen (tip). You find this (technically) almost on every graphic tablet. The software needs a threshold (or an adjustment of an existing one). (To an extreme, get a capacitive pen and try it on a mobile phone: you will see how the lines are miss aligned - probably not that jaggy, as the touch screen does a fair job of filtering the points)
Absolutely, but that is the issue here, the system isn't doing pretty much anything with the input data and 3 steps are kinda the key to have, especially with 4096 level of pressure sensitivity: mild low pass filtering, optimization of overlapping points, resampling of the curves and conversion to bezier. This can all be done as a post process, so it wouldn't affect the writing speed. And since the device is dual-core, it can be dumped onto the non-primary core, so it can do it in the background. Or at the absolute least, during export.
Either the 1/3 of users is not performing the test correctly (or are unwilling to admit there is a problem) or some devices used different internal components (most likely in connection to the digitizer driver) and they don't exhibit the issue. This is just thinking out loud, I have zero evidence to confirm this, so take it as such, please :)
@@MyDeepGuide Well. It's a mechanical / electrical problem with this spread. The digitizer glass, the pen tips,.. If they are not exact the same you get the spread.
Voja, great review once again. I recently received the “check your address” email from Remarkable so my device should be on its way soon, however I’m waiting for your Boox Air comparison before I decide whether to keep RM2 or replace with the air. Do you have a date when that comparison video will be ready? Thanks a lot
I got my remarkable 2 yesterday (batch 4), no signs of the jagged lines for now. The only way I can reproduce the issue is if I pess on the screen really hard with the pen when I'm writing!
Really good analysis, thanks. My RM2 have the same problem too: is there at least someone who don't have that problem? If so, in that case it's an hardware issue with some (many) RM2.
Hi Voya, as always I am enjoying your deep reviews. Since I am looking for opportunity to boost my productivity with all these E-ink gadgets, when will you have finally the new onyx note air as I really want to know if it matches the writing experience of the reMarkable - which one you would recommend?
I have received it yesterday, and the unboxing video has just been posted. In-depth review will come soon, hopefully next week, and after that also the comparison with the rM2.
I'm wondering whether the electromagnetic environment of the RM2 when it is tested could not explain the variability of the jagged lines issue between tablets/tests. Would it be possible to compare what happens in an EM free environment (far from AC wires/antennas including WiFi off, phones, microwaves,...) compared to an EM cluttered one (WiFi on, phone nearby, close to AC wires, microwave on, ... 😊)?
I've had my reMarkable for almost 3 weeks now and I'm not sure when to change the nib on the Marker. Do you have any suggestions on when is the best time to change it?
I am rather interested in how much irregularities exist between the devices (particularly the Lumi and Air) compared with the remarkable series. Did you have the opportunity to run the same test for the Boox devices perhaps?
Voja, wouldn't it make sense for the RM2 being twice as fast screen-wise, that during writing you also have twice the "frame rate" and also twice the "sampling rate" resulting in about twice as many vector points on the RM2 compared to the RM1? I hope it's solvable in the software.
Let‘s hope that this is not what they sacrificed to get the speed advantage over RM1. Otherwise a software fix (applying a low pass filter) might make it similar to the RM1. Accuracy and speed wise.
Is there any correlation as to whether one has this issue or not and batch number? If so this may focus part of the investigation on specific production runs. Someone with greater Facebook expertise than me may want to do a user group survey. Also, does the CANVAS display have substrate granularity, something like wood grain perhaps? It just seems curious to me that this is an issue with vertical lines on rM1 and horizontal lines on rM2. It’s almost as if reMarkable turned the original CANVAS display 90º and reformatted the aspect ratio to fit the rM2 form factor. Plus doubling the layers of pressure sensitivity to 4096 seems to quadruple the degree of SVG error.
For now, I am doing a basic tally of rM2 owners on FB, who has the issue or not, to get a rough ratio idea between affected and not affected units. Batch runs dependency is something reMarkable should do, and would need to do, if they are taking this seriously. So we'll see.
Hi Voya, thanks for the video. Quick question though, I went back to the original deed review you done earlier this year. On the brush test it doesn't show the same issue, you underlined all brush types as well. Can this issue be caused by a software update? Or do you think the test device didn't have it? Also you had R1 for a while I would say you would have noticed the issue on it as well without others pointing it out.
I have noticed some imperfections but I wasn't looking for it on the pre-production device of the rM2. As for the other question, it's a strange question and I'm not sure I fully understand it. Either way, I've had my rM1 for 3 years and I haven't noticed the issue, mainly because it's far less present and noticable on rM1 than it is on the affected units of rM2. I think I talked about that in the video and showed the examples, so people can see the differences. It's not a software update issue, as people are reporting having it on rM2 before update 2.3, so no. Also, some units are affected, some are not, unfortunately it seems that majority of units are affected. Either way, software fix is possible, we just have to wait and see what the Remarkable as a company will do about it.
Just found a very interesting compatibility issue. I am using a wireless charging mouse pad. And I found it will interfere with rm2's screen if I put it on top of the mouse pad and writing.
As always, a fantastic video... A thumbs up and subscribe doesn't do you justice... I wish there was some kind of machine-learning-driving indicator that evaluates intelligence of the host/content so I can filter my video feed and only get notifications for such quality content! Also love the way you perceive such issues. Too many sedated customers out there screaming for refunds whenever they fear "something is not right :-\"
Someone else has the problem that highlighting on PDFs doesn't mark correctly? I'm reading a bought programming pdf and when highlighting some text, it either doesn't mark where I said it should or it makes a freehand line.
Hmmm, nope I haven't experienced that, but I have read that since the last update, some users have experienced the highlighter reverting back to old, non automatic behavior.
Just came in. I also got the jagged lines. But it doesn't bother me that much. Maybe for critical sketching it would be an issue. In covid times, also I work mostly at home. I don't have a lot of paper at home, so Remarkable 2 is a handy tool. I was planning to only use it privately.
post processing requires clock cycles. They may need to put a timer on vertex point output, which would effect their writing speed, corner edge, etc... I would expect they are aware of the issue but have no immediate, meet the deadline, solution. Thanks for the video. I can relate with too many vectors points, drawing tree leaf's is crazy with too much data. I vote use point sampling, then from the start vertex, average the data to the end vertex. Let the vertex type be user selectable.
I have a question as I am not techy. Can it be assumed that if the vector data needs filtering or processing to smooth the line this will increase latancy as another process is required before the line appears question mark
Not at all. These processes are usually done after the initial input and are filtered "on the fly" (as the user is writing) or after the user has finished writing one continuous stroke (meaning as soon as you lift the pen off the screen to move to the place where you will start writing the next word) the device starts the optimization and post-processing of the last stroke. So no, if implemented normally, not at all.
I am wondering if this condition could be caused by the introduction magnetism into the remarkable 2? Which would make it a hardware problem and not software. What do you think?
I don't think so. This is a standard noise that is simply not being filtered at all and I suspect this is how they managed to get the speed out of the screen. Speed at the expense of quality.
It seems like depends on the device and how hard the person's writing habit is. On Reddit a user has already introduce a fix, however i don't have access to a Linux computer, so i can't test it, but it seems like it works great. If rM will do it? I don't know. You can try and ask them, people have done that and their reply was that they don't consider it a problem and that it was a calculated part of the "design" basically.
I'm a bit worried about my order now. I'll be a new reMarkable user; I don't have an RM1. I'm hoping that the right approach is to wait and see if it's a problem for me. I'm not much of an artist, so I hope it won't bother me too much... but it's hard to un-see a problem once you know it's there....
When I got the Remarkable 2 I noticed that my handwriting looked more jagged than on my Remarkable 1. I also have this issue with straight horizontal lines. :-(
Check the whole video, as I give out an advice in the end on what to do and how to do it in order to get remarkable to be aware of the issue and to raise it in priority, so that it gets solved sooner :)
@@MyDeepGuide I didn't have the 2.3 when I got the Remarkable 2 and I noticed the same jaggedness in handwriting from the first tests. The only fix we can hope for is software tweaks.
this i am going to check out tomorrow, but i have another issue, when i write my to do list, i write a date, always when i write for example 06.11.2020, the last 20 need longer time to appear, fells like a second. i also tried man different pen and marker, fine liner and so on, someone else have this delay issue?
Yeas, their official response is that it was a "balanced compromise" and since it doesn't bother majority of the users they will think of they can maybe look into it sometime in the future. Rather ridiculous response, imo. On another note, user Funkey has made a hack that does eliminate the issue and you can find the guide video on my channel, just search for it. But be aware that it is a hack and that you'll be doing it entirely at your own risk.
And here is a full grid: imgur.com/a/l32PQnk as one can see there is some warped spots where the screen is not calibrated well. And also the lines are not just jagged in horizontal directions but also in vertical at the edges. I used a ruler and held the marker plus at 0 angle
Holy moly... I'll do the same test on mine and on the other devices to see what I get and compare the results. This is quite an interesting thing to see where each of the devices stand. Thank you for this!
I believe you make some assumptions which are incorrect. You should not conflate the number of drawing data points with the "crookedness" or deviation from center of the same data points. Lots of data points equals higher resolution. This is why some displays have auto-smoothing applied to the datapoints. The error may be in how the touchscreen calculates those datapoints (it's a faster touchscreen input) or in some software difference. You also say it's not an software error as both RM1 and RM2 use the same software. I doubt both devices have binary identical versions of software. They may have identical version numbers, but they may be 2 different builds as they have 2 different processors and 2 different dispalys and 2 different touchscreen devices.
Obviously the "resolution" is higher, but the point, and the problem is that reMarkable 2 isn't filtering the data at all, which will of course result in a jumbled mess and unnecessarily large files. Raw input and high resolution is great, but it must be optimized and filtered, because in it's raw format it is expected that it's messy. Have you ever seen a gamepad or a joystick or even a mouse raw input axis? If games or 3D programs used RAW input data without the driver filtering them it would all be pretty much unusable for any kind of precision work. Same thing applies here :) As for the software, I was talking about the OS version not on a binary level (hence the air quotations), but rather that it's not something that was introduced with the OS 2.3 version, and this has been confirmed by users who noticed the issue prior to 2.3. Hope this clears it up a bit :)
@@MyDeepGuide Yes, I'm pretty familiar with 3D (check my TH-cam) and you make good points about filtering the data. Based upon your findings, the changes in hardware, the lack of occurences of the "crooked lines" in all devices, and my gut, I suspect the issue is in the touchscreen hardware. If that is the case, it could mean a long road for Remarkable, before it gets things square.
This has been an issue since the Remarkable 1. Not surprising. Even with a straight metal ruler I can not obtain a nice straight line. Gets close but not as good as I hoped.
Hi there, I have noticed that this jagged line problem doesn't appear on the blank page but it can be seen on some of the templates like the lined paper you have here. just a thought
I just tested, and I get it equally, regardless of the template used. But I discovered some new things, so I'll cover that at some point in time to see if that helps someone track down what the issue might be.
First thanks for your deep analysis. I hope they will fix it. I use RM1 and at the moment I think there is no reason for changing to RM2. It´s very expensiv and for that it should work perfect. My personal impresion is, that RM2 has just a smarter look, but it isn´t really better. Not sure if this is the right way.
I have tried 4 different markers with 4 different tips, well 3, since I used rM1 tip in the Boox pen :) This is shown all in the video, in case you missed it.
@@MyDeepGuide i'm still watching the video because i need to watch it at 0.75x because My english isn't good. But now i've see that you have tried different tips and marker. I'm sorry.
@@MyDeepGuide I don't remember if you tried to draw the lines after the 400% enlarged page. In my case in the page enlarged by 400% the line is better than the same drawn in the page 100%. Could this be a clue that the problem is software and not hardware?
In part and thanks to amazing Voja's reviews I've cancelled my RM2 preorder in favor of Onyx's latest release Boox Note 3. For a few bucks more I get a lot more functionality, storage, export/sharing options, etc... it's not all about paper & pen feel, just a personal preference of course. And no jagged lines whatsoever. Does anybody know why this company charges 400 regardless of the currency used which seems to be tied tied to the country of purchase/delivery? Sterling pounds, euros and US dollars are not worth the same so someone purchasing from UK and paying in pounds still pays £400 (eq $517 / €443), it doesn't sound too fair does it?
I haven't noticed on my end. I get pretty much the same things. The only thing that PDF and PNG formats have is a bit of a antialiasing, but the jaggedness remains, for me at least.
@@MyDeepGuide the zoom level also seems to change the effect, if you zoom right in and draw a straigt line it doesnt get all jagged. draw a line at 100% = jagged. zoom in to 500% and draw a line above it and it comes out perfectly smooth (for me at least)
If I understand correctly, both rM 1 and 2 run on the same software (same version, same firmware), but only rM2 experiences the jaggedness. Thus, the issue cannot come from the filtering, as the same filtering procedure is applied on both devices. My hasty conclusion would be that this problem is hardware-related, since this is, in principle, the only difference between both devices. Is my reasoning correct?
Not really, because even though the OS number and appearance are the same, they are different systems because they run on and run different hardware. So, under the hood, they aren't the same.
My understanding is that the issue is present on both rm1 and rm2, but more exaggerated on the rm2 and worse in the horizontal plane, whereas it's worse vertically on rm1.
if the position system is denser in the rmble2 than 1, it could be part of the difference. The fact that preassure effects the result might indicate it is effected by the hardware, ex the screen flexes in some way that effects how the position is recorded... If they decide to implement post-processing I sincerely hopes it will be togglable/adjustable how harsh the optimisation is. I would hate if something intensional got corrected!
is it effected by where on screen? that is if more or less jaggered if drawn close to edge of screen? (would indicate that some flexing is part of effect)
Since finding a way how to report an issue is something that's been made extremely difficult by the reMarkable on their website, I'm sharing a link here where apparently (as far as I can figure it out) we can fill out the form and send the ticket:
support.remarkable.com/hc/en-us/requests/new?ticket_form_id=360000022918
Thanks for checking this issue. It happens on mine as well, just reported it to support (although I'm not hopeful they will address it for months to come or that they will give any kind of timeline, given all past experiences).
Thanks. Just received mine (batch 3) an it also has the issue. Just reported it to support and asked for advice on what should I do either ask for a replacement or wait for a software update.
Ohhh, this is great, thanks! The more people we can get bugging them about stuff, the better our chances. Videos online from guys like you are even a bigger motivator to them, so thanks again for those too. Keep up the great work! 👍
Hi Voya thanks for the effort and honesty as always. I believe the question for me (and some others) is whether the device is still a better choice when compared to others (Boox air for example), especially for the ones who are waiting, since the selling point until now has been the writing experience - and if the writing is jagged I feel like the writing exp overall would not be that wonderful anymore, and together with the wait, the lack of some functionalities and tools - does it still give a run for its money?
@@NecroDomoEPI I have just sent the issue to them as well, along with the files I have exported and showed in the video, an explanation of the issue, a link to this video, and the report from the Facebook poll that out of 97 who responded 69 are affected by the issue, which is 71%. This is far worse than I initially thought because I thought that it would be the other way around.
So, we'll see if they respond and address this in any meaningful way. I hope that they do because this affects directly their main selling point and the main advantage of the device in relation to the competition. So it would be very unwise to ignore it, and not treat it with high urgency and importance.
The best thing we can all do is report the issue so many tickets turn into a software (post line smoothing) fix ... excellent uber detailed analysis as ever. Thanks!
Your attention to detail is remarkable.
Voya (I hope I got the spelling correct), you are unique in our field and a true public asset. Your technical knowhow is exemplary. However, your best asset is your advocacy persona. I work in medical research and often connect with patient advocates. Finding that balance of technical knowledge, clarity on how an issue affects different groups differentially, and passionate advocacy for the group as a whole is what makes and advocate a true public service. You have convinced me to support you and your work through the patron program as you are a vital public service. Thank you!
Thank you very much for your kind words and the support! This really means a lot :)
I'm not returning my Remarkable 2 for sure as it is very usable, but I agree with you that Remarkable must step up their software game. But this might be so simple as settings tweaks or it needs some filtering.
Or just a calibration, there are many possible fixes, I just hope they do it soon, coz now I can't use rM2 for one of the main purposes I needed it for :/
And do you have the issue or not?!
@@Nekotaku_TV I do
Voja you are amazing. All that scientific analysis AND a philosophy lesson as well. It's strange how hiding behind a keyboard gives riteous indignant types the license to 'blow off' a company over one bug. Your 'stay calm and just report the facts' was EXCELLENT advice in this and many other situations. Well done and also for saying that you Love the device (RM-2).
Thank you, this is very much appreciated :)
Yep. You are getting better and better at trying to control the ' final effect'. Keep up the good job!
I believe reMarkable would have been well aware of this behaviour, how could they not.
Fixing the issue was either going to make honouring delivery dates impossible, or would negatively affect the user experience so badly that they decided to go ahead with a compromised product and worry about it later..., maybe both!
I'm concerned that they won't see it as something that needs to be fixed at all, but just how their tablets work in order to achieve the low latency and paper like experience.
I hope reMarkable responds quickly and positively next week, otherwise I might just stick to real pens and ink for a while longer.
Many thanks again Voja. I much appreciate your honest and comprehensive reviews.
I honestly hope that's not the case.
Batch 8 here. Still haven't gotten it. But please, let us know any updates on the issue (answers, timelines, etc.).
I've grown worried about this and about the reading capabilities of the device... I'll wait for mine and probably fill in the ticket as soon as i get it.
Thanks for your work!
@Voja I had an idea directly, but I'am not so experienced in *svg.
Do you think the huge amount of point's and resulting splines are responsible maybe for the low max. page storage(7000 instead of claimed 100000). May you check how big the svg from 10 horizontal lines are on the rM2 vs. rM1(If not I understand and no comments needed).
The lag of storage was one reason why I cancelled my pre-order.
There are some insane compression algorithms on the market but maybe they are not using them or there own isn't good enough.
Great video and very interesting. In some cases you are really deep investigating, 10 times more than me.
I'am learning a lot from you, and even if I would/should do it the same, I often enough don't do it. It's really nice to watch your work and reflect about it.
Thank you for your efforts!
Voj. You are a legend. Thanks for taking the time to analyse this!
I also have Remarkable 1 and 2.
I have observed very mild horizontal jitter lines on R2 which I am able to virtually eliminate or make worse by very simple methods.
Please note, I am not suggesting this will work for everyone, rather it may give some clues to those with a better grasp of the physics of these pen/screen systems regarding what may be going on. YMMV.
As soon as I saw your posting I was reminded of the experiments I carried out on R1 with transparent (physical) drawing templates.
I was interested in whether I could cut designs, or geometric shapes, through thin transparent plastic sheets (like tech drawing templates) to speed up some drawing tasks.
I quickly realised the pen body could easily ride up on edges of cut out templates, curves or rules, unweighing the nib, which stopped drawing. I therefore moved on to clear sheets of Lexan that were not engraved right through but would guide the nib.
I observed there seemed to be an optimum thickness of template sheet for getting smooth lines. I found that as the height of the pen above the surface rose too high, the lines could become progressively more jagged, until they look like seismograph traces.
It is nonetheless interesting to draw on clear plastic a few mm thick and watch the lines appear well below the pen tip.
Incidentally, I have previously tried various other template materials and found interesting effects .Metal edges for example could displace lines and slots just wider than the nib in thin metal shim gave (non displaced) jaggies even when the nib was on the R1 surface.
Anyway, today I found using my R1 pen to rule lines on my R2 the jaggies (which are not very bad to start with) almost disappeared.
On inspection, I noticed the R1 pen nib was warn down, by a measured 0.21 mm. I swapped the nibs between pens and confirmed the short nib eliminated the jaggies on both pens when used on R2. I then got out the plastic sheets I had used for my previous experiments and was able to induce proportionally worsening jaggies on both R1 and R2 , with both pens/nibs, in any direction, by increasing the pen height above the surface. I did observe the reported relative horizontal/vertical line differences between the R1 and R2.
My observations suggest that depending on the angle of the pen, the height of the pen body above the surface can be quite critical, in relation to how many points get recorded while ruling a line on an axis.
Small variations in manufacturing tolerances may be summing or subtracting to account for whether the jaggies become annoying or not, in addition to whatever filtering or smoothing may be going on.
I will be interested to hear if others have any luck with worn nibs of varying lengths.
Very interesting And thank you for sharing :) it could be a thing, that the screen protector installed on top is inconsistent in a way. However, have you tested the pressure, because for me, it was the main thing that made it worse (regardless of the pressure or tilt sensitivity of the brush) pure physical pressure results in worse jaggies, for me and quite a few others as well.
@@MyDeepGuide Yes, I also found that increasing pressure gave more/bigger jaggies with the pens I have tried: fine liner, ballpoint, tech pencil and calligraphy nib.
I tend to use the fine line pen for most writing and diagrams, having previously enjoyed using Rotring tech drawing pens and /or 4B thin mechanical drawing pencils on drawing film.
Since posting above, I have experimented further.
I progressively filed down a nib from 2.88 mm (shoulder to tip) to 1.8 mm but did not see any marked improvement over the 0.2mm wear I started with. I think the R1 pen may be marginally less prone to jaggies but the difference is very small and only evident with the shorter nib.
I got the best looking horizontal lines drawing on R2 surface using the fine ballpoint nib, the 0.2 mm worn nib and a light touch. The magnitude of jitter shown was less than a quarter of line thickness.
I observed variation between one line and the next, however, which got me thinking about sources of noise and resonance and how they may interact with the various analogue to digital to digital conversions.
I am also involved in photography and have encountered interesting issues converting jpeg (raster images) into vector images, where matching sampling frequencies and adjusting filters can be crucial.
To explore some likely sources of noise leading to jitter.
I used the fine liner on R2 with R2 pen and a new nib, but I placed a variety of films on top of the R2 surface, intending to modify surface induced vibrations. I also varied pen pressure.
The results proved very interesting.
0.12mm, very smooth mylar sheet, with light pressure made almost perfectly smooth lines.
Increasing the pressure produced some intermittent "stiction" with occasional low-medium amplitude jiggles, depending on pressure, along with some low amplitude higher frequency noise.
The paper drawing film stuff (0.08 mm) the Remarkable came wrapped in, produced very uniform, meduim to high frequency, hi amplitude jaggies which varied less with pressure.
Textured 0.54 mm plastic sheet that has a similar to, but scaled up surface to the remarkable, produced similar results to the remarkable but with even lower amplitude, low spatial frequency, jaggies on light pressure and even higher amplitude on higher pressure.
As a result of these brief tests, I am inclined to think that the pen nib is picking up surface texture (including the intentional "paper like feel" vibrations on R screen surface) and/or frictional vibration.
Maybe due to mechanical resonance and/or sampling effects, these vibrations are being interpreted as horizontal movement, maybe due to the apparently high resolution of the sensor, to show up as jitter. Presumably there is some crucial difference in how vertical and horizontal movement is processed.
These ideas may explain why varying the pressure on the nib has such a big effect on level of jitter. Like the preload on a record player pickup cartridge.
So far, the magnitude of any problem is minimal for my everyday use of Remarkable.
I usually draw with a light touch in any case.
If I need to draw precise, straight horizontal lines, I can always lay down some Mylar sheet.
Indeed, my past experiments on R1 led me to make a thin Lexan ruler, with a smooth V groove engraved, almost all the way through, which constrains the pen nib and facilitates smooth straight lines.
I was able to do some very precise drawing by using the powerful 500% Zoom facility, so long as I stayed away from the edges where things get non linear.
I explored using the remarkable as a tech drawing input device by placing it on a mechanical drawing board. The idea was to convert the R1 input to vector then clean it up if needed on a computer. The intention being to use familiar, fast, tech drawing methods and physical tools to bridge the gap between a rough sketch and CAD. I explored making some mini mech drawing tools sized to the R 1, using linear digital scales and digital angle measuring tools. So far it still looks feasible, assuming the necessary desktop software conversion tools can be made to work reliably. I am keen to see what you and others may find there.
Like you, I think, hope, there may be some tweaks to the sampling/rendering process that might be useful in time.
So far I don't have the time to explore a vector export workflow in depth.
Similar issues may also occur in photography. There have been debates over anti aliasing filters on sensors vs post processing solutions and so far we seem to have moved in the direction of the latter.
Again I remain unsure whether my findings will generalise to others use cases.
Keep up the great work!
Thank you so much for doing this! When I started writing on my remarkable 2 I thought I was going crazy! I hope they come up with a fixed soon!
Batch 8 asking would you have canceled your order if you'd have known the issue beforehand?
Your channel truly lives up to its name. You've enabled the pickle to be both deeply ruminated by myself and pickled in the rumens of collective ruminants in the same stationary ecosystem. Let us anticipate the cultivation of patches with less messy pickles for the future.
Good analysis. Sensible approach to recognizing bugs do happen. I agree that this is an opportunity for the company to demonstrate their responsiveness. They have been extremely generous with people exchanging defective devices for good ones, so I believe they are capable of matching this on the software side. They do need to prove it to keep our goodwill, however.
Thanks for putting this on my radar, Voja! Otherwise I could've overlooked the issue in the trial period as I have finally received my device today, and it's my first e ink device. It's a truly beautiful piece of hardware but unfortunately I have the issue as well, although at first I thought I didn't. Two take aways:
1. It does indeed get worse if the pen is perpendicular to the screen.
2. Pressure adds to it.
3. It gets truly unworkable for me on the bottom cm of the screen, add to that that the line seems to be offset by 1 or 2mm in that same area.
I've sent a support ticket but I truly hope it can be fixed by calibration or a software fix, otherwise I'm almost certainly going to send back or sell the device. Probably the first as I don't want to resell a defective device. It's way to expensive for that.
It for a large part takes away the joy and excitement I had opening the package as it's uncertain if I can even use this in my daily routine.
Edit: batch 5
I wonder if this is the result of optimization for speed (lowering the latency in response to the pen) on the rM2 at the risk/sacrifice for smoothness in tracking the pen. That is to say, they may have changed the interpolation of points when a line segment is drawn from point A to B.
Also, just also wanted to add that given their build development environment you can still have the same OS version available for both devices, but you can have MACROS defined that would take advantage of the target hardware given that the rM2 has a slightly more "powerful" processor and double the memory.
Probably this is due to the fast screen reaction/responsivenes of the RM2. They would have to "fix" the line some seconds after you drawing, as filtering needs time and data from the "future" points. If doing so, you get the impression, that the line changes after drawing. That could be not so "paper like". So it is somehow a trade of. But for sure, that needs a lot of work software wise.
And: Thank you for your deep dive into that!
I been following you for like two years now and I swear you look younger and better with every video man!
Thank you very much :) I figured some stuff out, compared to last year, so I feel a lot better too :)
what kind of company allows product to go out their door like this ? I've been waiting 6+ months now for the rm2 and this excellent video just made me lose my cool. Sheesh! I feel like I've been robbed by an eloquent pick-pocket.
The issue is not present on all devices. Some units have it, some don't.
@@MyDeepGuide if those that don't have it REALLY don't have it then that points to a hardware issue. Maybe the installation of the screen is faulty ? I've seen reports of the surface of the screen protruding unevenly above the surface of the casing.
I don't have enough information to venture a guess on that.
I understand...great channel !
Thank you :)
Remarkable need to hire you!! You seem more passionate than people at the company
Oh I don’t want Voya muzzled by a corporate non-disclosure! He’s on our side! 😊
I'd like Remarkable teams start to improve his comunication.
It' ok have some issues, but it's their job not only fix issues, but even advice us about their plan for the next months.
Because if they don't want to share our future I'll send back my remarkable 4 when it arrives (batch 4).
Instead of sending it back you should sell it to me, as I'm on batch 8 :) I'm serious.
@@robmanzano7235 why I should sell you something that probably I'm gonna love it? 🙂
@@yoshi.jeje_StructuralEngineer hahaha, of course you won't 🙃
@@yoshi.jeje_StructuralEngineer Im batch 8 too, this waiting is killing me
Remarkable should hire you. You would design the best device ever
indeed
Voya, i'm very impress with the quality of your analyse. I'm waiting for my remarkable 2 (batch 4) and your video help me to wait. Hope that they resolved this problem soon.
apologies for my writing of your first name Voja.
HOLLY MOLLY thats a in-depth review .... thanks man for your work, really appreciate it. Made it possible to understand the product .
Wow thank you for your work!
Thanks Voja for the hard work.
Let's hope the remarkable team will fix that issue quickly and seriously (I'm in batch 8).
Thank you for watching :) Their response they sent out to users reporting on the issue (i haven't received a reply though for now 10 days, i guess they are no longer talking to me even to the support tickets), isn't the most encouraging one, as they basically downplay the issue and say that it doesn't "impact most users", even though it seems that the majority of the devices (or all) are affected. So yeah, I'm not sure they are taking it seriously, based on their response.
thank you for this, i considered buying it for mainly doodling on the go. in light of this, what would be your first choice for drawing on?
Thanks for this clear and deep analysis of the situation. It is understandable for a person like me who is not a technician 😂
I really appreciated the advice you give at the end telling people to report gently the problem, I do believe that there's no need to be angry in order to get satisfied 🙂
I'll keep my RM2 recieved yesterday since I'll use it essentially for writing, and I'll report the issue 😉
Only if you have it :) Since there are devices that do have the problem and some don't, so be sure to check :)
@@MyDeepGuide im sorry if I was unclear: I checked and I have the issue, yes 😅
@@MyDeepGuide But if this is really the case, wouldn't it make more sense to try to get a replacement? I mean ideally we would love to have better raw data than smoothend post-processed curves....
Even though I do not own a Remarkable 2, but I was very enjoy watching this video, you just eared one more subscriber.
You can test the issue on other devises and make a comparison. Like boox air, which adops the same panel.
Just did it on the Note Air for two software pen types, no issues.
No issue on the Max Lumi either.
I also love to see that comparison... So much talk about the quality of remarkable on writing and we get this?...
Is there also an svg export functionality in the native note app of boox? Thanks in advance.
@Long Johnson Sadly no - looks like \boox is exporting ONLY rasterized images (bitmaps) even in PDF format - its a major flaw !
My guess is that the screen refreshes and scans for inputs in a certain direction at certain intervalls. Due to the different form factors of the reMarkable 1 and 2 screens, those directions might be different (vertical on the R1 and horizontal on the R2). The reason the issue is present on the R1, but worse on the R2 might be caused by the higher "refresh rate", as in: lower pen latency of the R2.
So I reckon it might be the exact same root cause on reMarkable 1 and 2 and is actually a software issue that can be fixed via an update.
Absolutely a very likely culprit. It's also possible that 4096 levels of pressure sensitivity instead of 2048 come into play as well, because data is being piled up almost one on top of each other. Or a combination of both.
Either way, post processing optimization filter, especially on a dual core is a background process that shouldn't affect the performance of the device, yet would offer immensely better end results, not just for SVG, but for PDF and PNG exporting.
@@MyDeepGuide While post-processing would just be a band-aid for this issue, I agree that it's a good place to start as it addresses so many other issues, like file size and thus large notebook performance.
One would think that this kind of a conversation was supposed to be happening in Remarkable's R&D offices this time last year while they were developing the rM2, and not on a TH-cam channel, 2 months into shipping the units to users. Strange...
@@MyDeepGuide then wouldn't it be interesting to know if users that do NOT have the issue on their RM2 may just try vertical lines instead? Because _maybe_ the panel or digitizer manufacturing can be oriented both ways?
Also another thought: What if this was an intentional way of reducing the latency? I mean: more points per second would mean lower latency. But also more points, as we see now, also translate to fuzzy curves
@@faroit I'd bet money that people who don't experience this effect on their R2 have it with vertical lines instead.
It's probably not intentional, but the lower input latency amplifies the effect.
I was really looking into buying this, but after this I am done. Thank you! I am pissed off that they lied about Remarkable 1's marker not working on the 2nd generation. You can find it on their site. Cool idea, really wanted a writing only tablet, but... too bad.
@@beyondbased I understand that and I agree. The pressure sensitivity is important, but look at this:
"Does the Marker for reMarkable 1 work with reMarkable 2?
No, the Marker and Marker Signature for reMarkable 1 are not compatible with reMarkable 2.
The new Markers are calibrated to work with reMarkable 2’s second-generation CANVAS display"
It s a marketing scheme. The answer is valid, but the question is there to induce confusion.
I found this commwnt from facebook group. I can tell this guys knows what he is doing :)
Quat: "A long time ago, I worked on a high end commercial illustration program called Aldus FreeHand. It was the head to head competitor with Adobe Illustrator at the time until Adobe started gobbling up everyone. Part of that work included creating bezier curves to a set of points and also create outlines from bitmap images.
When I look at the jagged lines problem that shows up on RM2 but not RM1, I see that the RM2 is capturing points at a much higher rate than the RM1. These points also seem to have a higher "variation" of being to the left or right of the pen path than the RM1 did.
It makes sense to me that there are more points captured because the RM2 is running faster than the RM1.
The work I did converting bitmap images to lines tells me the jagged lines on the RM2 can be cleaned up with a software optimization. You can fit a straight line to that mess of points collected and you can also selectively eliminate some of the points collected.
What I don't know is how whether the hardware gives consistent results from one pen stroke to the next. If the results are farily consistent for any pen stroke (i.e. you get the same number of points and they create the same jaggy points), then it's a straightforward software fix. If the hardware is acting more randomly, you might still be able to fix the issue easily or it may require a bit more work.
Another fix is that RM can probably tune the rate at which it collects points for any given pen stroke. This will reduced the fidelity back toward RM1, but might be an easy quick solution."
It's a bit more complicated than that, and the best software solution would be 3 steps:
1. Gentle low pass filtering to take the wild peaks and put the jittering into a more equalized noise.
2. Point proximity optimization, which would determine the distance between points (any with any in a given stroke, not just consecutive ones) and weld together the overlapping points or those that are within a tiny and carefully calibrated threshold radius.
3. Finally, the most important step, sampling the optimuzed stroke and rebuilding it using bezier curves to get the closest possible match (can be pretty much perfect).
This is basically handing out the solution to the reMarkable team how to fix this on the secondary thread or as a regular coroutine as a post process after each stroke is completed. But they won't bother with it, that much is clear from their response.
Shoot! One of the main reasons I have the ReMarkable2 on order, and that is for drawing. I am so sick of sketch books everywhere.
I guess I'll have to see how bad it is when I get it. Thanks so much for the deep guide into this issue.
Great video, as always. Got my Remarkable 2 today, hopefully, mine does work properly, if not, hopefully, they will fix it. Thank you for all your content!!!
Very well done, as usual! Such a pragmatic approach.
I must confess, I can get a bit firey on the Facebook pages, and I think that's amplified by how much I 'want' to love this thing. I mean the hardware is just... Stunning!
I'm also keeping my preorder, and hoping to god that this is fixed soon 😁
I've said it before, but the hardware design team are basically saving this company right now!
Absolutely :)
I’m looking into purchasing the Remarkable 2, but after watching this video I’m starting to have a second thought.
You said that Remarkable 1 and 2 use the same software, but then you said that we don’t know if it’s a software issue or a hardware issue. However, if it’s not having much of an issue in Remarkable 1, then is it a hardware issue?
Also, this great study you’ve made was made a year ago, and I wonder if the issue been solved.
Thanks,
Jake
Hello,
This issue has since been resolved, because if you look at the video date, you'll see that it's quite old. But this is no longer a problem on the Remarkable 2.
I particularly appreciate the tutorial on how to behave when commenting or reporting issues. Thank you Voja.
(And... Nerdy things! 😊 The vector format check)
Thanks for posting this video showing in detail the problem. The issue is easy to replicate and so I have logged a ticket with remarkable too. I really hope this is fixed because it seems the pen feel of this device is unmatched by their competitors (even if the functionality is lacking in comparison).
Remarkable are pariahs. I'll test it out but there bs answer -no answer really took several weeks to get a response that was not a response. Love your content. Any thoughts on amount of magnets in device that could cause issues for mechanical watch?
Please guys fill the ticket, if we are only few of us they ll ignore it for sure. Btw thank you for your review I'm always impressed by how you do it well.
Those lines look to me a work of their communications group- jagged and irregular. If you know what I mean 😭
I bought a Remarkable 2 recently and its upto date and still has this issue, any fixes?
Thanks for the detailed look into the problem. It could be related to the faster response rate on the RM2. If there's less time to sample the location of the marker then the location accuracy will be lower. If that's the case it could easily be fixed in software by applying a little bit of filtering, though that may make writing small text more 'blunt'.
I have the issue, but as I use the RM2 mainly for notetaking it doesn't bother me. But I did notice trying it out is that there is a difference in the amount of jaggedness when you tilt the marker. When you tilt the marker while drawing a line the jaggedness is substantilly less.
One thing I notice, if I use the pen at a steeper angle, it has less jagged pieces. It does it I hold it more upright. Why, no idea.
Why they don’t put the option « ruler » like on the whiteboard microsoft ?
I wonder if the increased pressure causes neighboring sensors to trigger as well, registering additional "touches".
I'm having a similar issue with my Max 3 (although not that bad). I've been using my Max 3 for the past 2 months and love the device. I notice my traces are a bit shakey though. Someonw said it has to do with the inaccuracy of the EMR screen. Is there a setting I missed that can smooth out my traces when I write?
Second day of ownership for me. I tried and confirmed the jaggedness. It can be smoothed out to a degree by creating the lines in zoomed view. It is a workaround but does not require extra software
Hi there. Love the content. Has this jagged lines been resolved by Remarkable via an update? Is it even a software issue?
Yes it has, years and years ago...
Remarkable one had similar issues when writing near the edge... it would put a lip at the tops of letters when writing.
An unrelated question: how long does it take to fully charge RM2(e.g. battery from 8% to 100%)? It took me 8 hrs to get from 8% to 85%(not 100%). I am not sure if it is only me or it is usual that RM2 takes long time to be fully charged.
I’m 17 minutes into the video. I have to leave for work, but I wanted to share some initial thoughts:
1) Just because you have the same OS version between the rM1 & rM2, it does not mean that the two devices share the exact same digitizer hardware, digitizer firmware, and/or digitizer drivers.
I’ve seen this kind of issue in early android phones. The nexus one for example, could not draw straight lines very well at all, and the lines were very wavy.
This kind of issue could stem from badly designed or faulty digitizer hardware, or a bad digitizer firmware, or a bad digitizer driver. It’s also possible that different rM2 Devices are being shipped out with different digitizer hardware from different vendors.
22 minutes into it now. A couple of additional thoughts:
* You tried varying pressure, but not angle.
* Some people in the comments are confusing antialiasing (which I think the Boox devices do) with point smoothing (not a graphical process change, but just a change to the way the point data is interpreted). I wanted to clear that up, but I didn’t have time.
21:46 dang, that’s some crunchy data. It’s as if it was filtered through a guitar distortion amp 😂
Clearly, some corners were cut, either in the hardware, firmware, drivers, or software.
Yeah, this is beyond wavy lines and AA. It looks like the raw input is very garbled, especially the triangle, and yeah it can be anything at this point.
Post processing proximity filtering, optimization and smoothing can be a good temporary patch, helper thing until they figure out what's really going on, but evidently, some unites have this, some don't. Would be good to get a rough ratio estimate to understand better what's up.
I tried angle, but video was already long enough, no difference. That's also why I used the Fineliner, coz it's the bread and butter brush, no pressure, no angle.
@@MyDeepGuide
Fair ‘nuff, thanks Voja
I am pretty sure it was already mentioned, but this "jaggy" lines look like an issue between the higher sensitivity, the different surface and the capacity of the pen (tip). You find this (technically) almost on every graphic tablet. The software needs a threshold (or an adjustment of an existing one).
(To an extreme, get a capacitive pen and try it on a mobile phone: you will see how the lines are miss aligned - probably not that jaggy, as the touch screen does a fair job of filtering the points)
Absolutely, but that is the issue here, the system isn't doing pretty much anything with the input data and 3 steps are kinda the key to have, especially with 4096 level of pressure sensitivity: mild low pass filtering, optimization of overlapping points, resampling of the curves and conversion to bezier. This can all be done as a post process, so it wouldn't affect the writing speed. And since the device is dual-core, it can be dumped onto the non-primary core, so it can do it in the background. Or at the absolute least, during export.
Still, why if 1/3 of the devices are performing fine without filtering, those would like to be able to turn off any Post-Processing
Either the 1/3 of users is not performing the test correctly (or are unwilling to admit there is a problem) or some devices used different internal components (most likely in connection to the digitizer driver) and they don't exhibit the issue. This is just thinking out loud, I have zero evidence to confirm this, so take it as such, please :)
@@MyDeepGuide Well. It's a mechanical / electrical problem with this spread. The digitizer glass, the pen tips,.. If they are not exact the same you get the spread.
Voja, great review once again. I recently received the “check your address” email from Remarkable so my device should be on its way soon, however I’m waiting for your Boox Air comparison before I decide whether to keep RM2 or replace with the air. Do you have a date when that comparison video will be ready? Thanks a lot
Thank you, I don't have the device yet, it should arrive on Monday, so maybe in the first week of November.
@@MyDeepGuide many thanks Voja
I got my remarkable 2 yesterday (batch 4), no signs of the jagged lines for now. The only way I can reproduce the issue is if I pess on the screen really hard with the pen when I'm writing!
Will ReMarkable extend their 30day trial period until this is fixed?
I wouldn’t hold my breath on that.
Haha no.
Hi Voja! Sorry for asking this here, but will you have the opportunity to review the new Onyx Boox Note 3? Thanks. Great videos!
Really good analysis, thanks. My RM2 have the same problem too: is there at least someone who don't have that problem? If so, in that case it's an hardware issue with some (many) RM2.
Hi Voya, as always I am enjoying your deep reviews. Since I am looking for opportunity to boost my productivity with all these E-ink gadgets, when will you have finally the new onyx note air as I really want to know if it matches the writing experience of the reMarkable - which one you would recommend?
I have received it yesterday, and the unboxing video has just been posted. In-depth review will come soon, hopefully next week, and after that also the comparison with the rM2.
I'm wondering whether the electromagnetic environment of the RM2 when it is tested could not explain the variability of the jagged lines issue between tablets/tests. Would it be possible to compare what happens in an EM free environment (far from AC wires/antennas including WiFi off, phones, microwaves,...) compared to an EM cluttered one (WiFi on, phone nearby, close to AC wires, microwave on, ... 😊)?
Errrr... No :)
Remarkble if youre reading this- I ordered a unit and its on way. If this is the case it may not be useable for me since I plan to export to dwg.
I've had my reMarkable for almost 3 weeks now and I'm not sure when to change the nib on the Marker. Do you have any suggestions on when is the best time to change it?
I am rather interested in how much irregularities exist between the devices (particularly the Lumi and Air) compared with the remarkable series. Did you have the opportunity to run the same test for the Boox devices perhaps?
Boox devices have post processing and have had it for a long time now.
Voja, wouldn't it make sense for the RM2 being twice as fast screen-wise, that during writing you also have twice the "frame rate" and also twice the "sampling rate" resulting in about twice as many vector points on the RM2 compared to the RM1?
I hope it's solvable in the software.
It would, but it would also need to be addressed so that the end result is smooth and at least as good as rM1 and the competition.
Let‘s hope that this is not what they sacrificed to get the speed advantage over RM1. Otherwise a software fix (applying a low pass filter) might make it similar to the RM1. Accuracy and speed wise.
Is there any correlation as to whether one has this issue or not and batch number? If so this may focus part of the investigation on specific production runs. Someone with greater Facebook expertise than me may want to do a user group survey.
Also, does the CANVAS display have substrate granularity, something like wood grain perhaps? It just seems curious to me that this is an issue with vertical lines on rM1 and horizontal lines on rM2. It’s almost as if reMarkable turned the original CANVAS display 90º and reformatted the aspect ratio to fit the rM2 form factor. Plus doubling the layers of pressure sensitivity to 4096 seems to quadruple the degree of SVG error.
For now, I am doing a basic tally of rM2 owners on FB, who has the issue or not, to get a rough ratio idea between affected and not affected units. Batch runs dependency is something reMarkable should do, and would need to do, if they are taking this seriously. So we'll see.
Hi Voya, thanks for the video. Quick question though, I went back to the original deed review you done earlier this year. On the brush test it doesn't show the same issue, you underlined all brush types as well. Can this issue be caused by a software update? Or do you think the test device didn't have it? Also you had R1 for a while I would say you would have noticed the issue on it as well without others pointing it out.
I have noticed some imperfections but I wasn't looking for it on the pre-production device of the rM2. As for the other question, it's a strange question and I'm not sure I fully understand it. Either way, I've had my rM1 for 3 years and I haven't noticed the issue, mainly because it's far less present and noticable on rM1 than it is on the affected units of rM2. I think I talked about that in the video and showed the examples, so people can see the differences.
It's not a software update issue, as people are reporting having it on rM2 before update 2.3, so no. Also, some units are affected, some are not, unfortunately it seems that majority of units are affected.
Either way, software fix is possible, we just have to wait and see what the Remarkable as a company will do about it.
Just found a very interesting compatibility issue. I am using a wireless charging mouse pad. And I found it will interfere with rm2's screen if I put it on top of the mouse pad and writing.
As always, a fantastic video... A thumbs up and subscribe doesn't do you justice... I wish there was some kind of machine-learning-driving indicator that evaluates intelligence of the host/content so I can filter my video feed and only get notifications for such quality content!
Also love the way you perceive such issues. Too many sedated customers out there screaming for refunds whenever they fear "something is not right :-\"
I don't seem to have this problem on my device to the extend you show it.
Someone else has the problem that highlighting on PDFs doesn't mark correctly? I'm reading a bought programming pdf and when highlighting some text, it either doesn't mark where I said it should or it makes a freehand line.
Hmmm, nope I haven't experienced that, but I have read that since the last update, some users have experienced the highlighter reverting back to old, non automatic behavior.
Just came in.
I also got the jagged lines. But it doesn't bother me that much. Maybe for critical sketching it would be an issue.
In covid times, also I work mostly at home. I don't have a lot of paper at home, so Remarkable 2 is a handy tool. I was planning to only use it privately.
post processing requires clock cycles. They may need to put a timer on vertex point output, which would effect their writing speed, corner edge, etc... I would expect they are aware of the issue but have no immediate, meet the deadline, solution. Thanks for the video. I can relate with too many vectors points, drawing tree leaf's is crazy with too much data. I vote use point sampling, then from the start vertex, average the data to the end vertex. Let the vertex type be user selectable.
I have a question as I am not techy. Can it be assumed that if the vector data needs filtering or processing to smooth the line this will increase latancy as another process is required before the line appears question mark
Not at all. These processes are usually done after the initial input and are filtered "on the fly" (as the user is writing) or after the user has finished writing one continuous stroke (meaning as soon as you lift the pen off the screen to move to the place where you will start writing the next word) the device starts the optimization and post-processing of the last stroke. So no, if implemented normally, not at all.
Could it be that the improved response time of Rm2 has to do with the not-filtering? And that a software solution would increase the response time?
Not if it's implemented as a post processing filter, as it would happen after the strokes have been written.
I am wondering if this condition could be caused by the introduction magnetism into the remarkable 2? Which would make it a hardware problem and not software. What do you think?
I don't think so. This is a standard noise that is simply not being filtered at all and I suspect this is how they managed to get the speed out of the screen. Speed at the expense of quality.
@@MyDeepGuide Do you think there will be a software update to eventually fix the problem? Does it seriously effect the note taking aspect?
It seems like depends on the device and how hard the person's writing habit is. On Reddit a user has already introduce a fix, however i don't have access to a Linux computer, so i can't test it, but it seems like it works great.
If rM will do it? I don't know. You can try and ask them, people have done that and their reply was that they don't consider it a problem and that it was a calculated part of the "design" basically.
@@MyDeepGuide Hey can you link to that reddit post? I'm curious.
I'm a bit worried about my order now. I'll be a new reMarkable user; I don't have an RM1. I'm hoping that the right approach is to wait and see if it's a problem for me. I'm not much of an artist, so I hope it won't bother me too much... but it's hard to un-see a problem once you know it's there....
When I got the Remarkable 2 I noticed that my handwriting looked more jagged than on my Remarkable 1. I also have this issue with straight horizontal lines. :-(
Check the whole video, as I give out an advice in the end on what to do and how to do it in order to get remarkable to be aware of the issue and to raise it in priority, so that it gets solved sooner :)
@@MyDeepGuide I didn't have the 2.3 when I got the Remarkable 2 and I noticed the same jaggedness in handwriting from the first tests. The only fix we can hope for is software tweaks.
this i am going to check out tomorrow, but i have another issue, when i write my to do list, i write a date, always when i write for example 06.11.2020, the last 20 need longer time to appear, fells like a second. i also tried man different pen and marker, fine liner and so on, someone else have this delay issue?
I'm really dissapointed with this issue. It's the same on my reMarkable 2. I love the idea of this device. But it does not deliver what it promises.
Any news on this issue? I have similar problem on my but worse on vertical lines. Software version 2.5.0.27
Yeas, their official response is that it was a "balanced compromise" and since it doesn't bother majority of the users they will think of they can maybe look into it sometime in the future. Rather ridiculous response, imo.
On another note, user Funkey has made a hack that does eliminate the issue and you can find the guide video on my channel, just search for it. But be aware that it is a hack and that you'll be doing it entirely at your own risk.
@@MyDeepGuide Thanks!
@@MyDeepGuide I did the mod and it works perfectly. Big thanks!
@@MyDeepGuide to help other people find video quickly: th-cam.com/video/P770-5E-IFs/w-d-xo.html
And here is a full grid: imgur.com/a/l32PQnk
as one can see there is some warped spots where the screen is not calibrated well. And also the lines are not just jagged in horizontal directions but also in vertical at the edges.
I used a ruler and held the marker plus at 0 angle
Holy moly... I'll do the same test on mine and on the other devices to see what I get and compare the results. This is quite an interesting thing to see where each of the devices stand. Thank you for this!
you used a ruler? really???? WOW! are you sure you dont work for Onyx? ;-)
I believe you make some assumptions which are incorrect. You should not conflate the number of drawing data points with the "crookedness" or deviation from center of the same data points. Lots of data points equals higher resolution. This is why some displays have auto-smoothing applied to the datapoints. The error may be in how the touchscreen calculates those datapoints (it's a faster touchscreen input) or in some software difference.
You also say it's not an software error as both RM1 and RM2 use the same software. I doubt both devices have binary identical versions of software. They may have identical version numbers, but they may be 2 different builds as they have 2 different processors and 2 different dispalys and 2 different touchscreen devices.
Obviously the "resolution" is higher, but the point, and the problem is that reMarkable 2 isn't filtering the data at all, which will of course result in a jumbled mess and unnecessarily large files. Raw input and high resolution is great, but it must be optimized and filtered, because in it's raw format it is expected that it's messy. Have you ever seen a gamepad or a joystick or even a mouse raw input axis? If games or 3D programs used RAW input data without the driver filtering them it would all be pretty much unusable for any kind of precision work. Same thing applies here :)
As for the software, I was talking about the OS version not on a binary level (hence the air quotations), but rather that it's not something that was introduced with the OS 2.3 version, and this has been confirmed by users who noticed the issue prior to 2.3.
Hope this clears it up a bit :)
Or to put in 3D terms, unfiltered mocap data, have you ever worked with it? If yes, then you'll recognize the noise signature here as well :)
@@MyDeepGuide Yes, I'm pretty familiar with 3D (check my TH-cam) and you make good points about filtering the data. Based upon your findings, the changes in hardware, the lack of occurences of the "crooked lines" in all devices, and my gut, I suspect the issue is in the touchscreen hardware. If that is the case, it could mean a long road for Remarkable, before it gets things square.
Subbed, that's why I mentioned 3D references :) Thanks for all the awesome Blender content! :)
Would the addition of filtering impact the speed or responsiveness of writing or drawing?
Not if it's a post process
This has been an issue since the Remarkable 1. Not surprising. Even with a straight metal ruler I can not obtain a nice straight line. Gets close but not as good as I hoped.
Hi there, I have noticed that this jagged line problem doesn't appear on the blank page but it can be seen on some of the templates like the lined paper you have here. just a thought
I just tested, and I get it equally, regardless of the template used. But I discovered some new things, so I'll cover that at some point in time to see if that helps someone track down what the issue might be.
First thanks for your deep analysis. I hope they will fix it. I use RM1 and at the moment I think there is no reason for changing to RM2. It´s very expensiv and for that it should work perfect. My personal impresion is, that RM2 has just a smarter look, but it isn´t really better. Not sure if this is the right way.
I just received my RM2, and I have to say I'm quite unsatisfied with the amount of jaggedness in everything I write. It's really poor IMO.
Got batch 7, tincy bit but really no.
Sir u r the best means really best
Could it be a problem with the marker tip?
Have you try to remove the tip and put it again?
I have tried 4 different markers with 4 different tips, well 3, since I used rM1 tip in the Boox pen :) This is shown all in the video, in case you missed it.
@@MyDeepGuide i'm still watching the video because i need to watch it at 0.75x because
My english isn't good.
But now i've see that you have tried different tips and marker.
I'm sorry.
Not a problem at all :)
@@MyDeepGuide I don't remember if you tried to draw the lines after the 400% enlarged page.
In my case in the page enlarged by 400% the line is better than the same drawn in the page 100%.
Could this be a clue that the problem is software and not hardware?
In part and thanks to amazing Voja's reviews I've cancelled my RM2 preorder in favor of Onyx's latest release Boox Note 3. For a few bucks more I get a lot more functionality, storage, export/sharing options, etc... it's not all about paper & pen feel, just a personal preference of course. And no jagged lines whatsoever.
Does anybody know why this company charges 400 regardless of the currency used which seems to be tied tied to the country of purchase/delivery? Sterling pounds, euros and US dollars are not worth the same so someone purchasing from UK and paying in pounds still pays £400 (eq $517 / €443), it doesn't sound too fair does it?
When you export to pdf there seems to be some smoothing that happens to the jagged lines, is that a pdf specific thing do you think?
I haven't noticed on my end. I get pretty much the same things. The only thing that PDF and PNG formats have is a bit of a antialiasing, but the jaggedness remains, for me at least.
@@MyDeepGuide the zoom level also seems to change the effect, if you zoom right in and draw a straigt line it doesnt get all jagged. draw a line at 100% = jagged. zoom in to 500% and draw a line above it and it comes out perfectly smooth (for me at least)
@@fstirling3843 will it be annoying always having to zoom in and write QAQ
How fast does the pen tips wear out?
Keep up the good work ❤️
I´ve have only been reading bad news about this company lately: suscription models, delivery delays and now this news. I´m a little worried.
Shame, seems like a waste of genuinely innovative hardware
The worst thing is the increase of price from batch to match
Its a j I am sooke
A user received his/her Batch 3 product with an empty marker box. I saw this bad news on Raddit.
If I understand correctly, both rM 1 and 2 run on the same software (same version, same firmware), but only rM2 experiences the jaggedness. Thus, the issue cannot come from the filtering, as the same filtering procedure is applied on both devices. My hasty conclusion would be that this problem is hardware-related, since this is, in principle, the only difference between both devices. Is my reasoning correct?
Not really, because even though the OS number and appearance are the same, they are different systems because they run on and run different hardware. So, under the hood, they aren't the same.
@@MyDeepGuide Thanks for the clarification!
My understanding is that the issue is present on both rm1 and rm2, but more exaggerated on the rm2 and worse in the horizontal plane, whereas it's worse vertically on rm1.
if the position system is denser in the rmble2 than 1, it could be part of the difference. The fact that preassure effects the result might indicate it is effected by the hardware, ex the screen flexes in some way that effects how the position is recorded... If they decide to implement post-processing I sincerely hopes it will be togglable/adjustable how harsh the optimisation is. I would hate if something intensional got corrected!
is it effected by where on screen? that is if more or less jaggered if drawn close to edge of screen? (would indicate that some flexing is part of effect)
Maybe this is part of how they got the decrease in latency? Got rid of some filtering.
Kinda glad I cancelled my Batch 4 preorder. I'll just stick with my rM1 for now.