As a note about the vented backplate: 1. I tested VRMs, ICs, the SSD and the RAM for temperatures with the vented backplate and saw less than 1 degree difference from the stock backplate. 2. A lot of air still gets taken in through the rear vent, even with the vented backplate, so I have very few concerns with device longevity, but wanted to be fully transparent! Edit: I used the "sensors" command for SSD and motherboard temperatures, spot checked with the other results I gathered from MangoHUD. The ICs and VRMs were a bit more difficult. I used a temperature probe wired through the back vent to measure the surface temperatures, and verified its accuracy with a backplate-off IR thermometer test. Obviously the backplate-off results are cooler, but I verified that the probe was accurate that way.
Finally bro, thank you for relieving my stress with the vented back plate. I bought it but started to think about how tf air passively cools the Ics, ram, etc
I love this video, but I wish you went into as much detail about your validation of all critical component temperatures as you did with the rest of the testing methodology. The airflow the way Valve designed it is very much needed to keep things like the charging IC and MEM4 within safe temperatures, as per GN's analysis. While I am definitely open to the idea that the JSAUX backplate could be an improvement all around and would love the performance/aucoustic gains, I would need some more details than (crudely put, I am sorry) "trust me, I did some testing and it all seemed fine to me" in a comment. It seems like you actually did some thermocouple testing, I would have love it if that was included in the video! I'd love to modify my Deck in this way, but as it stands the details as to the thermal safety of doing so just aren't enough for me to risk my Deck.
Your videos are really cool! But what about the power consumption? If the SoC has more headroom in terms of temperature and therefore keeps the clock more stable, it should also consume a little more power, right? Or is that compensated by the lower power consumption of the slower rotating fan?
Kyle. You are the gift that just keeps on giving! There is a lot of scattered content on thermal modding but the video you created has the next level of polish to it. Sincere thank you from the community.
@@cryobyte33Hello cryobyte. I hope you're doing well. I've noticed that update 1.61 brought FSR 2.1 to cyberpunk 2077, so I was wondering if you were going to do a new performance video for the game?
I'd looked at thermal pads on Steam Deck to see if it was worth it, but never found any benchtest I could trust, but you went above and beyond as always! Seems like the backplate + thermal pad combo is absolutely worth it!
Thank you for watching! I'm just here to present the information, and I don't want to detract from it. Aside from trying to improve retention and make a few jokes, I'll be standing strong 😁
Thanks for the video! Some dB(A) info for you: - This measurement is on a logarithmic scale - Every 10 dBA is equivalent to a doubling of sound pressure. - Example: 2.7dBA is NOT 2.7x louder - that's closer to 30% louder IF measuring the exact same tone. - Example: an 80 dBA sound is 4X "louder" than a 60dBA one. This is the difference between a normal conversation face to face and a movie theater. - dBA is a measure of sound pressure - not necessarily perceived volume, and: - Higher frequencies will generally sound much louder to us than lower ones at the exact same dBA measurement.
I knew about the logarithmic scale and the frequency thing, but not the rest, thank you so much for explaining! I'll screenshot your comment and be sure to follow up with more research where needed, I appreciate it! ❤️
Thank you so much for the very kind comparison! I love Steve's work and I aspire to be similar in some ways, but then improve in others 😁 Hopefully I'll be able to get a little more than just Steam Deck content sometime too 😉
Thank you for these tests... Unlike a certain TH-camr that just screamed "higher temp bad" a while back, you did some great work looking into it. Amazing work!
Geebus this video was great. We had so my opinions and "I think this or that" in the community and now finally we have this glorious comprehensive test that told us a LOT. Thank you Cryo for another amazing set of tests.
Because of your video I also bought the Honeywell pad and the JSAux backplate. Its crazy how quiet the deck is now! Thank you so much! (Especially for cryotools!)
I just got my Vented JSAUX backplate today, and WOW the difference is instantly noticable! Any graphically intensive game would be loud as hell, but now it's super quiet, and GPU and CPU didn't go over 70c the whole time! Well worth the $30, was super easy to install too, highly recommended 👌 Thanks for this super informative video, never would have found this was it not for your awesome content 👍
Thank you so much for watching and for trying out the backplate! It's been a huge change in daily usability and I'm glad that you're getting the same results that I am 😁
To be honest, I don't follow this account, I've never seen another of his videos, I don't even have a Steam Deck, and I just accidently came across here. I'm strangely intrigued and thoroughly enjoyed the effort and thought process of this video. Keep up the good work dude!
Thanks for confirming that I made the right choice to replace the thermal paste with ptm7950 when I replaced my thumb sticks last month! Now I have a reason to get a backplate. 😁 Oh and I'm excited about the results of that monoblock! Keep up the good work!
Jsaux new backplate with thermal pad and fan intake vent looks promising especially if you want a quieter Steam Deck, but I'm worried it cools down the APU better but removes airflow to other critical components like SSD, ddr ram memory dies, voltage regulators, battery and other components that might get too hot now and fail earlier!
I mention that concern near the end of the video, but also mention that I watched all those temperatures and that they stayed identical and have given me no reason to worry about them 🙂 Thank you for watching, I'm glad it helped you out!
when you think about it... its much like a normal desktop pc right... the only components that typically get direct contact cooling with fan assistance or water cooling are the gpu and cpu with some motherboards getting a fan on the chipset but its often not required. everything else will have a smaller heat sink that is cooled passively, as long as that original opening is not blocked then even with the fan being open to get its air supply it should still drag air in from the original opening, i just dont see it being an issue, all components should get adequately passively cooled.
Would you consider making a follow up video on this when used in the long run? As you stated, there *could* be issues with removed airflow to other components, but it's probably impossible to tell in such a short amount of time. Would be interesting to see how or if it changes with time.
I might be able to later, but I might be doing other mods in the meantime so it'd be hard to isolate changes. What I can tell you, though, is that: 1. I tested VRMs, ICs, the SSD and the RAM for temperatures with the vented backplate and saw less than 1 degree difference from the stock backplate 2. A lot of air still gets taken in through the rear vent, even with the vented backplate. Hopefully this alleviates some of your concerns, thank you for watching!
@@cryobyte33 Thank you so much for taking time to respond! I ordered a backplate now, I understand it's at my own risk but I feel at ease now, regardless of how it goes
@@jhooGs Nope, not by any stretch of the imagination! You'd likely need a 8C+ difference over a long period, or 15C+ in the shorter term to make a significant difference of any sort.
Finally the Monoblock! I’m using the old jasaux backplate, new paste, monoblock and Fantastic from Decky. Temps are great with the custom fan curve. The monoblock with the back plate moves a ton of heat but boy does it get hot! Looking forward to the next video!
I have the OG JSAUX Backplate and I also did a repaste but after watching this video I bought the thermal pad as well as the new jsaux backplate. My temps on my OC Deck have been basically always in the 80s so seeing those temps drop will be greatly appreciated
It would be awesome if you did a Steam Deck OLED thermal mod including re-introducing thermal pads, thermal plate mod (to accommodate the pads contact), and PTM9750 APU instead of thermal paste, and the JSAUX back!
@@cryobyte33 the new thermal plate on the OLED unit has an insulating layer, and cut-outs in it for where the thermal pads from the heat pipe touch the metal plate. You’d have to cut the black insulation plastic layer wherever a thermal pad touches. The new OLED model doesn’t have thermal pads on most components now… It’s a cost cutting measure. You’ll see it if you look at the Gamer Nexus teardown to see what I mean.
I've had a conversation about theoriticals of this like an hour before you uploaded the video. I never expected to get my answers so fast. Thank you for the hard work!
I just stumbled on this video as I am waiting on my Deck to get delivered. Dude, this is an amazing video! It is by far the most complete and well thought out examination of the Deck's cooling solutions I have seen. Thank you so much for this! Its gonna take me a bit to muster the courage to go deep enough to change the thermal paste with the pad, but I have something to aim for. Keep up the great work!
I know you just did all this work and seriously thank you, but I'd love to see some of data for TG's new kryosheet also. That is what I'm running on both my ally and deck and from some rough testing it seems to be performing quite a bit better than stock pastes for both and even the kryonaut extreme I had on the deck prior. People like you shouldering the extreme time investment and hard work to not only do all the testing but consolidate and present it to the community really drives everything forward and makes everyone's lives easier. THANK YOU
Thank you so much for watching, and for the recommendation! I can't order any more products while I;m still here in Sweden, but I'll try to order it after I'm back in to US and do another cooling test video 😉 I'm really happy I was able to help a bit, hopefully it won't be too long before I can follow-up!
I recently fitted the 'old' jsaux back plate. I removed the thermal pad prior to installing the shell.. I wasn't after a shell to cool the deck down, more for the looks, since I opted for a transparent purple back plate. I don't notice any difference in temps... but now I feel that I maybe shouldn't of removed the thermal pad.. The reason I removed it, because I heard that the pad actually blocks the air from passing from the vent at the back to the fan.
Dude how do you not have over 1 million subs. I'm relatively new to your channel and from all I've watched. I've come up with nothing but amazing content. Plus, your download I installed on my deck is freaking sweet! Thank you man!
Just wanted to drop by and add my voice to the choir. I also changed to the Honeywell pad and jsaux backplate. This, together with a cryobyte-undervolt and a fan replacement changed my deck from “pretty damn annoying” to quiet and amazing. I’m extremely satisfied with these upgrades. The pad was a pain to install, but indeed worth the effort. Thnx for all your amazing guidance 👌
That warning for taking your morning coffee before working on your steam deck is REAL folks! I just did the JSAUX backplate mod last week and of course - forgot to remove my microSD card. It snapped in half leaving the rest of the card completely stuck in there and I only made it worse trying to use tweezers to remove. Luckily I have a 1TB SSD so won't be needing the microSD card slot...but a huge bummer nonetheless. Anyways - excellent video as always Kyle. Thank you!
This is a great video, but I had to pause on your charts quite often to read them. One thing that would aid their digestion is to color code. Since you often had 2 or 3 direct comparisons shown on the same chart, you could use related colors to distinguish the tiers. For example, you could use two shades of red for the overclocked comparison and two shades of blue for the non-over clocked comparison. Hope this helps and keep up the highly informative videos!
Thanks for the insights. I’d like to add 2 cents to it. The stock backplate isn’t that great in terms of thermal design. Warmer air has higher static pressure. Stock plate has intake vent right above the silicon cover, which is heating up the air and causing elevated pressure that will hinder cool air from flowing into the chasis. That’s why fan speed has to go much higher to counter that effect to keep air flowing into the chasis. I think the best backplate will be a vented cover with metal plate exposed for magnet peltier cooler.
another piece of top-notch content as usual! your dedication to thorough and fair testing is admirable, thank you so much for all of your hard work :) this info is very valuable and i'll definitely be considering purchasing at least one of these products in the future!
something i use that you may want to check out is a graphene thermal pad. it goes over the gpu and is pretty thin. beats putting paste way deep in there.
Amazing and thorough as always, super glad to see someone scientifically conduct testing of the PTM7950 with respect to burn in and thoughtful application. Cheers!
I'd like to see the temps of the charging IC and the SSD with the new Jsaux backplate. That extra cent hole probably hurts those way more than it helps the APU
I actually did test the VRMs, SSD, RAM and ICs (and mention it in the concerns section), they were within 1 degree of stock and gave me no reason to worry whatsoever 🙂 The fan still pulls air through the back vent and the metal plate also contacts the heatsink (which is getting better cooling). Thanks for watching, I hope this helps!
@@cryobyte33 apricides for video but I must ask - how you measured temps of IC, SSD, VRMs and battery? I heard yours concerns but I have mine as well th-cam.com/video/NeQH__XVa64/w-d-xo.html With this hole above fan I treat this backplate like there is no backplate at all because negative airflow is interrupted. I wonder what temps of battery are at your tests while gaming with docked deck.
@@RetroWaveArcades While negative airflow is interrupted, it still pulls a ton of air through the back plate. The fan can pull quite a bit more air than normally gets to it. As for the temperatures, I used the "sensors" command for SSD and motherboard temperatures, spot checked with the other results I gathered from MangoHUD, and checked. The ICs and VRMs were a bit more difficult. I used a thermocouple temperature sensor wired through the back vent to measure the surface temperatures, and verified its accuracy with a backplate-off IR thermometer test. GN does really good work, and I want to make it clear that I'm not trying to argue with Steve's results. All I know is that I did the best I could to to verify that this would be safe for long-term use. 🙂
Your long term concerns deserve to be reviewed at a later date maybe a year or two from now or even periodically? I think viewers would be interested in such content. Thanks for your hard work!
I might be able to do that, but I'd need a separate Deck to do it. As it stands, this Deck will be modified repeatedly for other testing so I can't do a "set and forget" follow-up 😓
@@cryobyte33 Thanks for the reply! I noticed that your video did not have general guidelines (i.e. in the introduction section) about what temperatures we should aim to keep the Steam Deck at. Noticed mine getting to around 85 C after playing for several hours and worried about what that temperature range (80-90C) might do to my Deck's longevity. I changed the CPU TDP under Performance settings to 10 and temps went down to 70-80C range. I will check if you have a video on optimizing settings in the Deck for minimizing temps.
@@manaseeker2414 The Deck APU is safe up until 105C. Since we were working with Stock temps or lower, I didn't think it was necessary to include that 😅 I mentioned the upper limits in my overclocking video, where I thought it was more relevant, sorry!
Man, dont think ive ever done this before but ...subscribed, hit the bell, dropped a like and even commenting. Amazing content. Thank you for the time you dedicated putting this all together.
Wow, that was one technical video that I don't normally watch and didn't really need to watch, but watched it to the end. Just love the Deck and what it can do.....Gaming has come a long way since the ZX 48K spectrum where I started. Love the utility by the way.
This is exactly what I've been wanting to see. Jsaux posted tests that only covered the internal temps of the APU, where my main concern was the components not cooled with the heat pipe.
I have the Delta fan but never found it bothering me since I always cap the fps/ screen refresh. I finally changed the backplate (vented) and thermal paste to NT-H1 and what a difference in noise! The fan is barely audible most of the time. I wish there weren't so many fake PTM7950 sellers out there. I'm happy with the Noctua paste for now. Thank you for your hard work!
Audio engineer here. With the acoustic measurements in the A weighting (which refers to how the volume of sound reacts to what we here, as we hear different frequencies at different perceived volumes), a 3dB difference is twice the acoustic energy - audio is measured on a logarithmic scale so it would be perceived as a small increase in sound, but the "2.7dB increase," for example, is not 2.7 times louder. It is less than twice as loud, but would be perceived as a small increase in volume.
The burst speeds make sense. Better thermal contact with the heatsink creates better coupling, resulting in APU temps closer mirroring the regular heat soaking of the heatsink.
For direct die application, one paste that for some reason people don't bring up that often, but it's likely one of the very best, is the Prolimatech PK-3.
Thank you so much for the kind words, and for watching! I think the Deck is great because we can do so much with it, and it's my favorite device as a result 😁
Again, an amazing and incredibly helpful video! Thank you so much! If it's not too basic for your channel, I'd absolutely love to see a video about how to find the perfect settings for any game on Steam Deck to guarantee steady FPS, maximise battery life while still having a decent gaming experience. I do know the basics but I have a hard time finding playing around with all the options the Steam Deck provides. Maybe with an explanation of what which feature does so viewers can easily transfer this to any game they want to play. Keep up your amazing work!
Saw you said you’re still learning acoustics basics. A lot could be said, but rough guidelines for you: 3 dB change is a just noticeable difference, 5 dB is clearly noticeable, and 10 dB is a doubling of loudness. But your points about different fans having different loudness or annoyance even with similar overall levels is great!
Thank you so much for letting me know! It's a big field of study that I hope to become competent at over time and every comment helps me get there. I was able to detect a 2dBA change, but I assume that's just because I was listening for it, fascinating stuff 😁
Installed the vented backplate on my delta fan deck, and also added the clip on fan. Deck runs cooler and the little leg on the clipon fan does work well as a stand. It's not fallen over, surprisingly.
Just got a 512gb OLED SD and have been loving your channel. I can't wait to hear your thoughts and modding recommendations on the OLED deck. Keep up the great work.
Hey Kyle, I appreciate your response - thank you! Two questions regarding your video: - With what tool / app did you set your Fan on the highest Speed after the 2 hour benchmark ? - How do you run the Superposition benchmark on the Steam Deck, is there a special version for the deck? Bests, Marc
You rock! This is super helpful information. I am stock right now but despite the tentative concerns on the new jsaux backplate, I am tempted to go that route with the PTM solution as well. I may never fully OC my deck, but I'm tempted knowing the headroom those two upgrade paths provide. Thanks for the work that you do, and I wish I could afford to support you on Patreon.
Great video! I didn't know jsaux made a new backplate. I just received and installed the old one about an hour before this video came and temps are night and day difference for me. Before 20w overclock would instantly go to 95C within seconds and maxed fans. Now after 10 mins it hovers around 89-90C.
Couple points: 1. I'd be mindful of using percentages when talking about temperatures, especially when you're comparing the raw output values. Worth delving into the thermodynamics of it if you see value in that. 2. Pulling cooling air away from the other components may be good for extending the life of the APU, but IIRC some of those VRMs are already running a bit toasty, so I submit that in doing so one will bring about the premature failure of a unit with a perfectly serviceable APU. Degradation of the main chip isn't the only way to kill a system. I would however love to see what results you get using Arctic MX-6. It's worked a treat on two laptops for me so far, so I'd love to see if it's worth doing up my deck with it.
Thank you for watching! 1. Good point, I'll try to stick deltas in there at least a bit. I didn't consider that since I typically talk about frame rates 🙂 2. I actually have a pinned comment about this, I tested that too and only noticed a 1C difference on the VRMs, ICs, etc. As for Arctic, those pastes tend to work best with high thermal loads whereas the Noctuas do better at lower loads. At least when I compared them on a few years ago they were identical up to about 45w of TDP, then Arctic pulled ahead. Thank you for the comment!
Kyle, another great video. I'd like to share my setup, experience and also ask a couple questions. First, I am running your CryoUtilties and followed your excellent Undervolt & Overclock video. My best stable setup was UV = 40/30/40: OC = CPU 3950 / GPU 1800 : TDP = 17 Watts. After replacing my back plate with the newest JSAUX, my current stable setup is UV = 40/20/40: OC = CPU 4300 / GPU 2100: TDP = 20 Watts. The back plate made a monster difference in deck temps. Second, I've used a few different tools for benchmarking & stress testing (cpu-x, Superposition, gtkstresstesting (locks up on me) & mprime) however, I've never seen my CPU run @ more than 4058 or my GPU @ over 2000. So, it doesn't seem to matter if I tweak the BIOS settings above 4000 & 2000 respectively. My next step is to crank up TDP to see if it makes a difference. Any thoughts on a better tools for benchmarking or what I may be doing wrong? Temps at my current config are low 70's to low 80's while pushing the deck hard, seems there's room for more juice.
In terms of the clip on fan from Jsaux which is only compatible with the stock backplat, I purchased two different versions of this, there is a kickstand built into the foot of the clip on fan which allows you to stand it. Also, there is a version which does not require the clip from Jsaux and is sold seperately as part of the ModCase, but it includes a metal stick on connector that also gives you the latch from the ModCase which lets you add other ModCase components (but not at the same time). I also tried PTM7950 and that was a total bust, in fact in some cases it made my Steam Deck hotter and definately made the standard case back a lot warmer. Maybe it depends on what model of Steam Deck you are testing on and if Valve used different paste in construction.
Brilliance, as always!!! Thank you for digging in and going the extra mile(s) on this (and all of your content, for that matter)!!! While I’m not in the market for any of these items, I love seeing the analysis. Also enjoying all the playful post-production touches, keep up the great work!
I'm just glad to put solid numbers out there for those interested. It's hard to find real data that isn't published by a specific company 🙂 I spent a LOT more time on editing this time around, so I'm really happy some have noticed!
great video! I have the old jsaux backplate. I slapped a blue anodized heatsink on it with a thermal pad and nothing gets too hot to touch. I have not overclocked. Sometimes the fans don't even run. That is when accessible parts get the warmest, but not hot enough to trigger the fan. I can even feel warmed air coming from the top vent on pure convection sometimes .
I have to say that this video was top-notch. I don't have a steam deck, but I watched as it showcased PTM7950, which is interesting as I'm looking to use it in a laptop and wanted to see what the performance would be like It's even better that you overclocked the stemdeck up to 20 watts tdp.
JSAUX back plate is no Joke, my wife's steam deck used to thermal throttle and just froze the games FULLY if the game was demanding. I used the JSAUX back plate, the one without the vent and changed the Thermal paste, added a thermal pad to the aluminum cover inside and BOOM, the temps were SOOO MUCH cooler. This is a must mod for the STEAM deck.
I didn't have the same issue with stock, but I agree that the JSAUX backplates are both big wins for thermals 😁 Thank you for watching and sharing your experience!
Watched this video twice. And i think i'll go with the thermal pad and backplate. If I can cut down the noise when playing in bed while the wife tries to sleep, then it's absolutely worth it. Thanks a lot !
I still haven't bought my SD, but will in a month. These videos are gold for enthusiasts like me who like to min-max every bit of hardware they got. I've already bookmarked the products you've linked and will be buying them first thing. Thanks a lot!
The clip opens up on the bottom! I do use the clip on fan and actually don't notice the noise, because I use it (kind of docked on its weird footpad) and wearing headphones or external speakers, that highly outperform the fan noise (Steelseries Arena 7)
Wow, I'm... not sure how I didn't notice that during testing 😂 Thanks for the reality check. I typically use my Deck in handheld without headphones, or with open-backed headphones, so the noise bothers me, but I'm glad it's working well for you. Thanks for watching!
I bought the new jsaux, but was on the hunt 4 the first gen with metal plate. After seeing this video. God im happy i went for the new. Really good video 😁
I need more guidance about putting thermal pads on the steam deck since I can't follow the process due to editing (and because I'm a noob 😅) did you put a thermal pad on the valve chip too? An installation guide would be helpful 🙏 Which size did you buy those thermal pad?
I put the pad only on the heat spreader, which the contacts the APU that you're speaking of. It's really easy, just treat it like thermal paste using iFixIt's guide 🙂 Thank you for watching!
Interesting to see you're using generous dimensions for the PTM pad. I've seen others cutting them to (exactly?) 13mm^2, which would be the size of the die. If you're thinking what I'm thinking, the pad is non-conductive and any excess size would only help transfer heat from the pad to the heatsink slightly. In any case, thanks for the extensive testing! :)
I ordered the pad, going to keep using original back plate for now. If nothing happens while using vent plate then I'll upgrade to it. Thanks for all your hard work.
Awesome video! This may be mentioned in another comment but for the fan volume measurements using a different scale for the dB might give a better visual representation of the perceived loudness. Especially for people not used to sound pressure level measurements or nonlinear measurements.
It seems I can't win 😅 In the last video where I measured dB I got a ton of complaints saying it should be dBA for more accuracy to the human ear. I got a meter capable of reading that and used it here for that reason 😂
@@cryobyte33 Sorry for the confusion the measurement technique is great. There are just conversions from db to sound intensity or perceived loudness that you can use. But you are right you can't please everyone haha either way thanks for the great content.
This is so cool, thank you for having the JSAUX. I have both, although I noticed that JSAUX might have taken into account one of my emails to them, because the RGB backplate (still hasn't arrived, so I can't say for sure), looks like, according to the images on their website, to have a section cut-out on the thermal pad to allow for the original Valve fluid dynamics diagram. If true, it will help a lot. I have also got a cheap (from Amazon) aluminium heatsink (40x40x11) with thermal adhesive to directly put over the aluminium backplate. This setup is perfect to use with the JSAUX also new RGB hub. I know it sounds like an ad lol, but I am just really excited for it to arrive (September according to their website).
Thank you for watching! I love JSAUX products thus far and don't have many complaints. I don't have an RGB plate, but hope to get one after I'm in the US to test.
I'm glad you addressed the potential issue with the vented rear plate. People generally think that more airflow is better, but almost always ignore airflow paths and how stock configurations are designed for the most efficient cooling all round, not just for a single component like the SOC. The SSD and VRMs will almost certainly be suffering worse temperatures with the vent over the fan. If you are able to see VRM and SSD temps, do you think you could do a follow up to confirm/debunk this?
So, I actually mentioned it briefly, but: 1. I tested VRMs, ICs, the SSD and the RAM for temperatures with the vented backplate and saw less than 1 degree difference from the stock backplate 2. A lot of air still gets taken in through the rear vent, even with the vented backplate, so I have very few concerns with device longevity. It's still a concern, but not nearly as much as I originally thought. I tested the vented backplate first because I was convinced that it would be bad for thermals, but I couldn't have been more incorrect, at least as of all my current findings. Thank you so much for watching!
@@cryobyte33 Ah I must have missed that part. If you're happy then I'm happy. I have the metal plate version and was thinking about changing it for the new one alongside a front plate swap, but had concerns so I'm glad you did the testing.
Thank you for this video. I have been on a cooling odyssee ever since i bought the deck myself. I tried out all the methods you demostrate in this vid, its really good to have you coming up with pretty much the same results. Cheers mate, keep up the awesomeness. Edit: With that being said, the weirdest thing: I spent half my time in a desert country, andwhen I'm there the old JSAUX backplate combined with a magnetic fan is what worked out best for me. Not so when I am back in western europe though. Also, there are concerns with JSAUX new backplate in regards to internal airflow and cooling of other elements, specifically the SSD. I even think to remember that Valve had once mentioned waz back that this was the reason for the absence of a second vent in the stock configuration.
Thanks for watching and for sharing your results! The locational difference is almost certainly humidity, with drier air you'd need more fan/watt to keep temperatures down. As for the new backplate: 1. I tested VRMs, ICs, the SSD and the RAM for temperatures with the vented backplate and saw less than 1 degree difference from the stock backplate. 2. A lot of air still gets taken in through the rear vent, even with the vented backplate, so I have very few concerns with device longevity. Hopefully this helps 😁
I absolutely love your content man. No other content creator I know of does such extensive and useful testing as you do for the steam deck. Keep up the great work as always, and I wish you the best going forward 😁.
Finally the new jsaux backplate reviews are coming out. Thanks so much for your detailed videos. I'm testing out a devaso backplate that has a metal plate attachment for a thermoelectric fan. Similiar to the jsaux magnetic fan but bigger
I’m curious as to how the KingPin paste would perform. Very thorough and informative video, thank you! I’ll be keeping the stock backplate but I think I’ll move up to the thermal pad 👍🏼
Great video, watched the entire thing, I do have a comment about the backplate though. I'd be interested to know if it could be resealed, how it would perform then in comparison to the stock plate. The negative pressure you point out, according to Gamer's Nexus, that allows area to be pulled over the battery circuit which if it doesn't do that, may be long term implications and could cause damage due to running at higher, possibly out of spec temperatures. If the newer backplate's, fan vent, was sealed, you may get the best of both worlds, and address the dust and noise concerns. Worth considering for sure - looking forward to the block testing :)
While dropping ~5C is definitely awesome. This just goes to show you how well designed the Steam Deck is from the get go. Valve really has some amazing engineers on this product.
I was thinking about getting the PTM7950 and new JSAUX backplate, but was concerned what the actual games were. This video definitely helps make my decision easier. I might just get the backplate and use some thermal paste I already have. Don’t want to setup the PTM7950 if it is all going to come out for a p3r. FYI I saw Salocin900 posted a video of an updated Steam Deck which has different mounting screw locations for the heat pipe. Might be an issue for the p3r if Valve has significantly changed the design.
Thanks a lot for your extensive testing and the detailed video, earned my subscription :) I got a few questions if you don't mind: 1. About the airflow from the original vent to the other components (VRMs, ICs, SSD, RAM), even though in your testing there was little variation in their temperature with the Jsaux vented backplate, there's definitely less airflow there, so would you say the additional airflow pull from the Jsaux clip-on fan compensates for that? I'm concerned about long term component longevity with less airflow. If not, would just closing that additional vent with tape make it like the stock backplate or is there more to it? 2. In the new vented backplate, the metal plate is not exposed like the old one. Is there nevertheless a significant advantage attaching a magnetic fan to to it instead of the metal plate directly? 3. Without overclocking, will I get significant performance advantages by thermal modding the Steam Deck? Thank you!
Thank you so much for watching and subscribing! 1. The clip-on fan SHOULD be able to compensate, especially since it would cover the vented hole, but the tape trick would work too. That said, I've been using mine OC'd with the JSAUX vented backplate for 3 months now with no issues! 2. The new backplate has an aluminum plate rather than steel, so it won't hold the magnetic fan. It also has much less thermal mass and a smaller thermal pad, so the magnetic fan would have a much smaller effect than the old plate. 3. You won't likely see performance advantages without overclocking, but you would have much quieter fans and marginally better battery life as a result! Hopefully this helps, thank you again for watching!
Seeing you’re video hit my sub box always puts the biggest smile on my face because i know how much hard work and time it takes testing things like this take, And someone with OCD&HDHD I really appreciate the effort you put into your testing with all the different independent variables there is when it comes to testing the smallest of things. I’m planning on water cooling my steam deck in an upcoming video if you ever have the free time I have some questions about attending to dd this stupid idea of mine
Thank you so much for watching the videos and being subscribed! I'm happy that you enjoy them as much as you do, and that it can hold your attention since both my wife and I have ADHD. As for water cooling, feel free to either email me at my business email, or join the Discord server and start a thread to chat through it 😁
Are there any updates on your concerns with cooling? I have the backplate myself and have had it about 3 months now. No issues on my end but i am by no means running testing on it.
@@cryobyte33 Thanks for responding! I was gonna ask the same question to see if your SD spontaneously combusted or something. Based on responses from Reddit, they make it seem like all the other components will die due to increased heat. They love saying how valve did negative pressure for a reason etc.
@@cryobyte33 thank you so much for replying. Your work is greatly appreciated. Huge fan here. I also have had no issues and the lower temps have been awesome.
As a note about the vented backplate:
1. I tested VRMs, ICs, the SSD and the RAM for temperatures with the vented backplate and saw less than 1 degree difference from the stock backplate.
2. A lot of air still gets taken in through the rear vent, even with the vented backplate, so I have very few concerns with device longevity, but wanted to be fully transparent!
Edit:
I used the "sensors" command for SSD and motherboard temperatures, spot checked with the other results I gathered from MangoHUD.
The ICs and VRMs were a bit more difficult. I used a temperature probe wired through the back vent to measure the surface temperatures, and verified its accuracy with a backplate-off IR thermometer test. Obviously the backplate-off results are cooler, but I verified that the probe was accurate that way.
OK that's the answer I was looking for :)
Finally bro, thank you for relieving my stress with the vented back plate. I bought it but started to think about how tf air passively cools the Ics, ram, etc
I love this video, but I wish you went into as much detail about your validation of all critical component temperatures as you did with the rest of the testing methodology. The airflow the way Valve designed it is very much needed to keep things like the charging IC and MEM4 within safe temperatures, as per GN's analysis. While I am definitely open to the idea that the JSAUX backplate could be an improvement all around and would love the performance/aucoustic gains, I would need some more details than (crudely put, I am sorry) "trust me, I did some testing and it all seemed fine to me" in a comment. It seems like you actually did some thermocouple testing, I would have love it if that was included in the video! I'd love to modify my Deck in this way, but as it stands the details as to the thermal safety of doing so just aren't enough for me to risk my Deck.
👍 Thanks!
Your videos are really cool! But what about the power consumption? If the SoC has more headroom in terms of temperature and therefore keeps the clock more stable, it should also consume a little more power, right? Or is that compensated by the lower power consumption of the slower rotating fan?
Kyle. You are the gift that just keeps on giving! There is a lot of scattered content on thermal modding but the video you created has the next level of polish to it. Sincere thank you from the community.
Thank you so much for watching another video! 😁 As always, I'm just happy to help.
I just had an idea. Drilling out vent holes in the back of the old Jsaux case would give you the best of both worlds right?
@@jackinabackpack7599 theoretically similar, but with a lot of the heat sunk into the backplate still 🙂
Kyle, you're a legend!
Thank you so much for being subscribed and watching the video! I hope it helps out!
@@cryobyte33Hello cryobyte. I hope you're doing well. I've noticed that update 1.61 brought FSR 2.1 to cyberpunk 2077, so I was wondering if you were going to do a new performance video for the game?
@@TheRealPaul_Morphy I plan to make an update after phantom liberty releases, since they're bumping the requirements 🙂
I'd looked at thermal pads on Steam Deck to see if it was worth it, but never found any benchtest I could trust, but you went above and beyond as always! Seems like the backplate + thermal pad combo is absolutely worth it!
100% but even good paste with back plate is worth it.
That's exactly why I wanted to make this video, thank you for watching it! 😁
Which backplate though? Stock or JSAUX?
@@rohacha9iin40 juasx v2
No flashy edits, no useless padded infos, just the good stuff. Never change Kyle.
Thank you for watching! I'm just here to present the information, and I don't want to detract from it. Aside from trying to improve retention and make a few jokes, I'll be standing strong 😁
Thanks for the video! Some dB(A) info for you:
- This measurement is on a logarithmic scale - Every 10 dBA is equivalent to a doubling of sound pressure.
- Example: 2.7dBA is NOT 2.7x louder - that's closer to 30% louder IF measuring the exact same tone.
- Example: an 80 dBA sound is 4X "louder" than a 60dBA one. This is the difference between a normal conversation face to face and a movie theater.
- dBA is a measure of sound pressure - not necessarily perceived volume, and:
- Higher frequencies will generally sound much louder to us than lower ones at the exact same dBA measurement.
I knew about the logarithmic scale and the frequency thing, but not the rest, thank you so much for explaining! I'll screenshot your comment and be sure to follow up with more research where needed, I appreciate it! ❤️
Watching this made me think I've seen this type of presentation. Then it hit me.
YOU ARE THE GAMERSNEXUS OF STEAMDECK
Thank you so much for the very kind comparison! I love Steve's work and I aspire to be similar in some ways, but then improve in others 😁
Hopefully I'll be able to get a little more than just Steam Deck content sometime too 😉
Tech jesus 2.0!!! 👊
Thank you for these tests... Unlike a certain TH-camr that just screamed "higher temp bad" a while back, you did some great work looking into it.
Amazing work!
Geebus this video was great. We had so my opinions and "I think this or that" in the community and now finally we have this glorious comprehensive test that told us a LOT. Thank you Cryo for another amazing set of tests.
Thank you for the kind words! I also noticed that there wasn't anything really solid out there so I wanted to do my part 🙂
Because of your video I also bought the Honeywell pad and the JSAux backplate. Its crazy how quiet the deck is now! Thank you so much! (Especially for cryotools!)
I just got my Vented JSAUX backplate today, and WOW the difference is instantly noticable! Any graphically intensive game would be loud as hell, but now it's super quiet, and GPU and CPU didn't go over 70c the whole time! Well worth the $30, was super easy to install too, highly recommended 👌 Thanks for this super informative video, never would have found this was it not for your awesome content 👍
Thank you so much for watching and for trying out the backplate! It's been a huge change in daily usability and I'm glad that you're getting the same results that I am 😁
To be honest, I don't follow this account, I've never seen another of his videos, I don't even have a Steam Deck, and I just accidently came across here. I'm strangely intrigued and thoroughly enjoyed the effort and thought process of this video. Keep up the good work dude!
Thanks for confirming that I made the right choice to replace the thermal paste with ptm7950 when I replaced my thumb sticks last month! Now I have a reason to get a backplate. 😁 Oh and I'm excited about the results of that monoblock! Keep up the good work!
Thank you for watching, I can't wait to test it and share the results! 😁
Jsaux new backplate with thermal pad and fan intake vent looks promising especially if you want a quieter Steam Deck, but I'm worried it cools down the APU better but removes airflow to other critical components like SSD, ddr ram memory dies, voltage regulators, battery and other components that might get too hot now and fail earlier!
I mention that concern near the end of the video, but also mention that I watched all those temperatures and that they stayed identical and have given me no reason to worry about them 🙂
Thank you for watching, I'm glad it helped you out!
wondering if covering the circular fan intake with duct tape does anything to performance and temps of the SOC.@@cryobyte33
when you think about it... its much like a normal desktop pc right... the only components that typically get direct contact cooling with fan assistance or water cooling are the gpu and cpu with some motherboards getting a fan on the chipset but its often not required. everything else will have a smaller heat sink that is cooled passively, as long as that original opening is not blocked then even with the fan being open to get its air supply it should still drag air in from the original opening, i just dont see it being an issue, all components should get adequately passively cooled.
I'm curious how loud ur deck runs. I have no issues whatsoever with fan noise
Would you consider making a follow up video on this when used in the long run? As you stated, there *could* be issues with removed airflow to other components, but it's probably impossible to tell in such a short amount of time. Would be interesting to see how or if it changes with time.
I might be able to later, but I might be doing other mods in the meantime so it'd be hard to isolate changes.
What I can tell you, though, is that:
1. I tested VRMs, ICs, the SSD and the RAM for temperatures with the vented backplate and saw less than 1 degree difference from the stock backplate
2. A lot of air still gets taken in through the rear vent, even with the vented backplate.
Hopefully this alleviates some of your concerns, thank you for watching!
@@cryobyte33 Thank you so much for taking time to respond! I ordered a backplate now, I understand it's at my own risk but I feel at ease now, regardless of how it goes
@@cryobyte33 would this less than 1 degree difference in internal components be enough to cause some kind of damage?
@@jhooGs Nope, not by any stretch of the imagination! You'd likely need a 8C+ difference over a long period, or 15C+ in the shorter term to make a significant difference of any sort.
Finally the Monoblock! I’m using the old jasaux backplate, new paste, monoblock and Fantastic from Decky. Temps are great with the custom fan curve. The monoblock with the back plate moves a ton of heat but boy does it get hot! Looking forward to the next video!
I can't wait to test it, thank you so much for watching!
I have the OG JSAUX Backplate and I also did a repaste but after watching this video I bought the thermal pad as well as the new jsaux backplate. My temps on my OC Deck have been basically always in the 80s so seeing those temps drop will be greatly appreciated
Agreed, the stock thermal system is great but not quite up to a good OC.
Thank you for watching!
It would be awesome if you did a Steam Deck OLED thermal mod including re-introducing thermal pads, thermal plate mod (to accommodate the pads contact), and PTM9750 APU instead of thermal paste, and the JSAUX back!
I haven't heard of the thermal plate mod, could you send an email with a link to the address on my about page? Thank you!
@@cryobyte33 the new thermal plate on the OLED unit has an insulating layer, and cut-outs in it for where the thermal pads from the heat pipe touch the metal plate. You’d have to cut the black insulation plastic layer wherever a thermal pad touches. The new OLED model doesn’t have thermal pads on most components now… It’s a cost cutting measure. You’ll see it if you look at the Gamer Nexus teardown to see what I mean.
🔥 hope you're settling in well mate, loving the content!
Haven't moved back yet, but thank you!
Jesus i literally was looking into this myself. Thanks man for always releasing that perfect video at the right time
Thank you so much for watching! I'm glad I could help a bit 😁
His name isn't Jesus.
I've had a conversation about theoriticals of this like an hour before you uploaded the video. I never expected to get my answers so fast. Thank you for the hard work!
Thank you so much for watching, and I'm happy to put some researched information out there!
New Deck owner and new arrival to your channel. Thanks for all your hard work and pleasant demeanor Kyle!
Thank you for watching, I hope the videos help out!
Have ordered the JSAUX new backplate, thanks for taking the time to test and also compile the info.
Thank you for watching, I hope you like it as much as I have!
I just stumbled on this video as I am waiting on my Deck to get delivered. Dude, this is an amazing video! It is by far the most complete and well thought out examination of the Deck's cooling solutions I have seen. Thank you so much for this! Its gonna take me a bit to muster the courage to go deep enough to change the thermal paste with the pad, but I have something to aim for. Keep up the great work!
I know you just did all this work and seriously thank you, but I'd love to see some of data for TG's new kryosheet also. That is what I'm running on both my ally and deck and from some rough testing it seems to be performing quite a bit better than stock pastes for both and even the kryonaut extreme I had on the deck prior. People like you shouldering the extreme time investment and hard work to not only do all the testing but consolidate and present it to the community really drives everything forward and makes everyone's lives easier. THANK YOU
Thank you so much for watching, and for the recommendation! I can't order any more products while I;m still here in Sweden, but I'll try to order it after I'm back in to US and do another cooling test video 😉
I'm really happy I was able to help a bit, hopefully it won't be too long before I can follow-up!
I recently fitted the 'old' jsaux back plate. I removed the thermal pad prior to installing the shell.. I wasn't after a shell to cool the deck down, more for the looks, since I opted for a transparent purple back plate.
I don't notice any difference in temps... but now I feel that I maybe shouldn't of removed the thermal pad..
The reason I removed it, because I heard that the pad actually blocks the air from passing from the vent at the back to the fan.
Dude how do you not have over 1 million subs. I'm relatively new to your channel and from all I've watched. I've come up with nothing but amazing content. Plus, your download I installed on my deck is freaking sweet! Thank you man!
Thank you so much for the kind words! I'm trying to get there, hopefully some day 🤞
Oh, and thank you for using CU as well!
Just wanted to drop by and add my voice to the choir. I also changed to the Honeywell pad and jsaux backplate. This, together with a cryobyte-undervolt and a fan replacement changed my deck from “pretty damn annoying” to quiet and amazing. I’m extremely satisfied with these upgrades.
The pad was a pain to install, but indeed worth the effort.
Thnx for all your amazing guidance 👌
What fan did you get? Also, what is the honeywell thing..?
That warning for taking your morning coffee before working on your steam deck is REAL folks! I just did the JSAUX backplate mod last week and of course - forgot to remove my microSD card. It snapped in half leaving the rest of the card completely stuck in there and I only made it worse trying to use tweezers to remove.
Luckily I have a 1TB SSD so won't be needing the microSD card slot...but a huge bummer nonetheless.
Anyways - excellent video as always Kyle. Thank you!
That's insane. You are already creating material at the Gamer Nexus level.
I hope all the best with the channel and the new life in the US.
Thank you so much for watching, and for the kind words! I hope to continue getting better and expanding the channel 😁
This is a great video, but I had to pause on your charts quite often to read them. One thing that would aid their digestion is to color code. Since you often had 2 or 3 direct comparisons shown on the same chart, you could use related colors to distinguish the tiers. For example, you could use two shades of red for the overclocked comparison and two shades of blue for the non-over clocked comparison. Hope this helps and keep up the highly informative videos!
A lot of information gathered and collated, you can tell this video took time and effort, have a thumbs up 👍
Thanks for the insights. I’d like to add 2 cents to it. The stock backplate isn’t that great in terms of thermal design. Warmer air has higher static pressure. Stock plate has intake vent right above the silicon cover, which is heating up the air and causing elevated pressure that will hinder cool air from flowing into the chasis. That’s why fan speed has to go much higher to counter that effect to keep air flowing into the chasis. I think the best backplate will be a vented cover with metal plate exposed for magnet peltier cooler.
The most comprehensive cooling video to date. Thank you for sharing all of this information!
Thank you so much for watching, I hope you liked it!
another piece of top-notch content as usual! your dedication to thorough and fair testing is admirable, thank you so much for all of your hard work :) this info is very valuable and i'll definitely be considering purchasing at least one of these products in the future!
Thank you for watching, I'm happy to have helped a bit and I hope you at least know all the implications of whichever choice you go with 🙂
something i use that you may want to check out is a graphene thermal pad. it goes over the gpu and is pretty thin. beats putting paste way deep in there.
Thanks for the recommendation, too! I've had one before, but I'll need to wait until I'm back in the US to get another. We'll see 😉
Amazing and thorough as always, super glad to see someone scientifically conduct testing of the PTM7950 with respect to burn in and thoughtful application. Cheers!
I'd like to see the temps of the charging IC and the SSD with the new Jsaux backplate. That extra cent hole probably hurts those way more than it helps the APU
I actually did test the VRMs, SSD, RAM and ICs (and mention it in the concerns section), they were within 1 degree of stock and gave me no reason to worry whatsoever 🙂 The fan still pulls air through the back vent and the metal plate also contacts the heatsink (which is getting better cooling).
Thanks for watching, I hope this helps!
@@cryobyte33 apricides for video but I must ask - how you measured temps of IC, SSD, VRMs and battery? I heard yours concerns but I have mine as well
th-cam.com/video/NeQH__XVa64/w-d-xo.html
With this hole above fan I treat this backplate like there is no backplate at all because negative airflow is interrupted. I wonder what temps of battery are at your tests while gaming with docked deck.
@@RetroWaveArcades While negative airflow is interrupted, it still pulls a ton of air through the back plate. The fan can pull quite a bit more air than normally gets to it.
As for the temperatures, I used the "sensors" command for SSD and motherboard temperatures, spot checked with the other results I gathered from MangoHUD, and checked.
The ICs and VRMs were a bit more difficult. I used a thermocouple temperature sensor wired through the back vent to measure the surface temperatures, and verified its accuracy with a backplate-off IR thermometer test.
GN does really good work, and I want to make it clear that I'm not trying to argue with Steve's results. All I know is that I did the best I could to to verify that this would be safe for long-term use. 🙂
Your long term concerns deserve to be reviewed at a later date maybe a year or two from now or even periodically? I think viewers would be interested in such content. Thanks for your hard work!
I might be able to do that, but I'd need a separate Deck to do it. As it stands, this Deck will be modified repeatedly for other testing so I can't do a "set and forget" follow-up 😓
@@cryobyte33 Thanks for the reply! I noticed that your video did not have general guidelines (i.e. in the introduction section) about what temperatures we should aim to keep the Steam Deck at.
Noticed mine getting to around 85 C after playing for several hours and worried about what that temperature range (80-90C) might do to my Deck's longevity.
I changed the CPU TDP under Performance settings to 10 and temps went down to 70-80C range.
I will check if you have a video on optimizing settings in the Deck for minimizing temps.
@@manaseeker2414 The Deck APU is safe up until 105C. Since we were working with Stock temps or lower, I didn't think it was necessary to include that 😅
I mentioned the upper limits in my overclocking video, where I thought it was more relevant, sorry!
Man, dont think ive ever done this before but ...subscribed, hit the bell, dropped a like and even commenting. Amazing content. Thank you for the time you dedicated putting this all together.
Wow, that was one technical video that I don't normally watch and didn't really need to watch, but watched it to the end. Just love the Deck and what it can do.....Gaming has come a long way since the ZX 48K spectrum where I started. Love the utility by the way.
This is exactly what I've been wanting to see. Jsaux posted tests that only covered the internal temps of the APU, where my main concern was the components not cooled with the heat pipe.
Thank you for watching! I'm glad that I could provide solid third party results 😁
Removing bottlenecks on my steam deck? Within reasonable price without replacing a huge amount of interior hardware? You got yourself a sub my friend!
I have the Delta fan but never found it bothering me since I always cap the fps/ screen refresh. I finally changed the backplate (vented) and thermal paste to NT-H1 and what a difference in noise! The fan is barely audible most of the time. I wish there weren't so many fake PTM7950 sellers out there. I'm happy with the Noctua paste for now. Thank you for your hard work!
Audio engineer here.
With the acoustic measurements in the A weighting (which refers to how the volume of sound reacts to what we here, as we hear different frequencies at different perceived volumes), a 3dB difference is twice the acoustic energy - audio is measured on a logarithmic scale so it would be perceived as a small increase in sound, but the "2.7dB increase," for example, is not 2.7 times louder. It is less than twice as loud, but would be perceived as a small increase in volume.
Thank you so much for the explanation, I clearly have some more learning to do 🙂
The burst speeds make sense. Better thermal contact with the heatsink creates better coupling, resulting in APU temps closer mirroring the regular heat soaking of the heatsink.
For direct die application, one paste that for some reason people don't bring up that often, but it's likely one of the very best, is the Prolimatech PK-3.
I actually haven't heard of that! I'll need to check it out when I get back to the US, thanks for watching and for the recommendation 🙂
LEGENDARY. AMAZING WORK KYLE. steam deck optimization is a very hot topic. No pun intended
Thank you so much for the kind words, and for watching! I think the Deck is great because we can do so much with it, and it's my favorite device as a result 😁
Just want to thank you for being thorough with this video. Really appreciate the time and effort you put in!
I was geeking out over this and I enjoy the detailed testing that you do.
Again, an amazing and incredibly helpful video! Thank you so much!
If it's not too basic for your channel, I'd absolutely love to see a video about how to find the perfect settings for any game on Steam Deck to guarantee steady FPS, maximise battery life while still having a decent gaming experience. I do know the basics but I have a hard time finding playing around with all the options the Steam Deck provides. Maybe with an explanation of what which feature does so viewers can easily transfer this to any game they want to play.
Keep up your amazing work!
Saw you said you’re still learning acoustics basics. A lot could be said, but rough guidelines for you: 3 dB change is a just noticeable difference, 5 dB is clearly noticeable, and 10 dB is a doubling of loudness. But your points about different fans having different loudness or annoyance even with similar overall levels is great!
Thank you so much for letting me know! It's a big field of study that I hope to become competent at over time and every comment helps me get there.
I was able to detect a 2dBA change, but I assume that's just because I was listening for it, fascinating stuff 😁
Great video! While delivering lots of data it's tricky to balance droning on forever and keeping the video moving. You nailed it. Huge thanks!
Incredible break down! Prepping for an eventual Deck purchase and this channel is a gold mine.
Installed the vented backplate on my delta fan deck, and also added the clip on fan. Deck runs cooler and the little leg on the clipon fan does work well as a stand. It's not fallen over, surprisingly.
Just got a 512gb OLED SD and have been loving your channel. I can't wait to hear your thoughts and modding recommendations on the OLED deck. Keep up the great work.
Hey Kyle,
I appreciate your response - thank you!
Two questions regarding your video:
- With what tool / app did you set your Fan on the highest Speed after the 2 hour benchmark ?
- How do you run the Superposition benchmark on the Steam Deck, is there a special version for the deck?
Bests,
Marc
You rock! This is super helpful information. I am stock right now but despite the tentative concerns on the new jsaux backplate, I am tempted to go that route with the PTM solution as well. I may never fully OC my deck, but I'm tempted knowing the headroom those two upgrade paths provide. Thanks for the work that you do, and I wish I could afford to support you on Patreon.
Great video! I didn't know jsaux made a new backplate. I just received and installed the old one about an hour before this video came and temps are night and day difference for me. Before 20w overclock would instantly go to 95C within seconds and maxed fans. Now after 10 mins it hovers around 89-90C.
They didn't advertise it very well, but I'm really happy they did. If the old one wasn't so hot it'd be easier to recommend 😅
Thanks for watching!
Couple points:
1. I'd be mindful of using percentages when talking about temperatures, especially when you're comparing the raw output values. Worth delving into the thermodynamics of it if you see value in that.
2. Pulling cooling air away from the other components may be good for extending the life of the APU, but IIRC some of those VRMs are already running a bit toasty, so I submit that in doing so one will bring about the premature failure of a unit with a perfectly serviceable APU. Degradation of the main chip isn't the only way to kill a system.
I would however love to see what results you get using Arctic MX-6. It's worked a treat on two laptops for me so far, so I'd love to see if it's worth doing up my deck with it.
Thank you for watching!
1. Good point, I'll try to stick deltas in there at least a bit. I didn't consider that since I typically talk about frame rates 🙂
2. I actually have a pinned comment about this, I tested that too and only noticed a 1C difference on the VRMs, ICs, etc.
As for Arctic, those pastes tend to work best with high thermal loads whereas the Noctuas do better at lower loads. At least when I compared them on a few years ago they were identical up to about 45w of TDP, then Arctic pulled ahead.
Thank you for the comment!
Kyle, another great video. I'd like to share my setup, experience and also ask a couple questions.
First, I am running your CryoUtilties and followed your excellent Undervolt & Overclock video. My best stable setup was UV = 40/30/40: OC = CPU 3950 / GPU 1800 : TDP = 17 Watts. After replacing my back plate with the newest JSAUX, my current stable setup is UV = 40/20/40: OC = CPU 4300 / GPU 2100: TDP = 20 Watts. The back plate made a monster difference in deck temps.
Second, I've used a few different tools for benchmarking & stress testing (cpu-x, Superposition, gtkstresstesting (locks up on me) & mprime) however, I've never seen my CPU run @ more than 4058 or my GPU @ over 2000. So, it doesn't seem to matter if I tweak the BIOS settings above 4000 & 2000 respectively. My next step is to crank up TDP to see if it makes a difference. Any thoughts on a better tools for benchmarking or what I may be doing wrong? Temps at my current config are low 70's to low 80's while pushing the deck hard, seems there's room for more juice.
In terms of the clip on fan from Jsaux which is only compatible with the stock backplat, I purchased two different versions of this, there is a kickstand built into the foot of the clip on fan which allows you to stand it. Also, there is a version which does not require the clip from Jsaux and is sold seperately as part of the ModCase, but it includes a metal stick on connector that also gives you the latch from the ModCase which lets you add other ModCase components (but not at the same time). I also tried PTM7950 and that was a total bust, in fact in some cases it made my Steam Deck hotter and definately made the standard case back a lot warmer. Maybe it depends on what model of Steam Deck you are testing on and if Valve used different paste in construction.
Brilliance, as always!!! Thank you for digging in and going the extra mile(s) on this (and all of your content, for that matter)!!! While I’m not in the market for any of these items, I love seeing the analysis. Also enjoying all the playful post-production touches, keep up the great work!
I'm just glad to put solid numbers out there for those interested. It's hard to find real data that isn't published by a specific company 🙂
I spent a LOT more time on editing this time around, so I'm really happy some have noticed!
great video!
I have the old jsaux backplate. I slapped a blue anodized heatsink on it with a thermal pad and nothing gets too hot to touch. I have not overclocked. Sometimes the fans don't even run. That is when accessible parts get the warmest, but not hot enough to trigger the fan. I can even feel warmed air coming from the top vent on pure convection sometimes .
how is it right now? still good? what about the dust build up as well?
I have to say that this video was top-notch. I don't have a steam deck, but I watched as it showcased PTM7950, which is interesting as I'm looking to use it in a laptop and wanted to see what the performance would be like It's even better that you overclocked the stemdeck up to 20 watts tdp.
I bought the Honeywell pad and installed it on my GPD win 4.
Glad these tests are further proof it actually is doing its job.
JSAUX back plate is no Joke, my wife's steam deck used to thermal throttle and just froze the games FULLY if the game was demanding. I used the JSAUX back plate, the one without the vent and changed the Thermal paste, added a thermal pad to the aluminum cover inside and BOOM, the temps were SOOO MUCH cooler. This is a must mod for the STEAM deck.
I didn't have the same issue with stock, but I agree that the JSAUX backplates are both big wins for thermals 😁
Thank you for watching and sharing your experience!
Excited for the Monoblock video! It will be the deciding factor in modding my Deck.
Excellent video, the data really is helpful to inform my purchase. Nobody else is making these kind of videos, huge props to you.
Thank you so much for the kind words and for watching! I'm just happy to help 🙂
Watched this video twice. And i think i'll go with the thermal pad and backplate.
If I can cut down the noise when playing in bed while the wife tries to sleep, then it's absolutely worth it.
Thanks a lot !
I still haven't bought my SD, but will in a month. These videos are gold for enthusiasts like me who like to min-max every bit of hardware they got. I've already bookmarked the products you've linked and will be buying them first thing. Thanks a lot!
Didn’t skip a single second. Definitely subscribing after this!
Great Work Kyle! Keep it up. Was a bless to watch and match my Modding Experiences ;)
Thank you so much for watching, I'm glad to have helped out!
The clip opens up on the bottom! I do use the clip on fan and actually don't notice the noise, because I use it (kind of docked on its weird footpad) and wearing headphones or external speakers, that highly outperform the fan noise (Steelseries Arena 7)
Wow, I'm... not sure how I didn't notice that during testing 😂 Thanks for the reality check.
I typically use my Deck in handheld without headphones, or with open-backed headphones, so the noise bothers me, but I'm glad it's working well for you. Thanks for watching!
Hope you give us an update about the vented backplate in the future.
I bought the new jsaux, but was on the hunt 4 the first gen with metal plate. After seeing this video. God im happy i went for the new. Really good video 😁
This is impossible to be more clear. Congratulations 🎉
I need more guidance about putting thermal pads on the steam deck since I can't follow the process due to editing (and because I'm a noob 😅) did you put a thermal pad on the valve chip too? An installation guide would be helpful 🙏
Which size did you buy those thermal pad?
I put the pad only on the heat spreader, which the contacts the APU that you're speaking of. It's really easy, just treat it like thermal paste using iFixIt's guide 🙂
Thank you for watching!
Need that clip on fan. Honestly I had one of those jet airplane ps4s so I’m sure the noise won’t be anything crazy lol.
Interesting to see you're using generous dimensions for the PTM pad. I've seen others cutting them to (exactly?) 13mm^2, which would be the size of the die. If you're thinking what I'm thinking, the pad is non-conductive and any excess size would only help transfer heat from the pad to the heatsink slightly. In any case, thanks for the extensive testing! :)
I ordered the pad, going to keep using original back plate for now. If nothing happens while using vent plate then I'll upgrade to it. Thanks for all your hard work.
very detailed and thoughtfully done. I'm thinking about the Noctua Paste and the Honeywell heat pad for sure.
dude you´re the best, you´re the bestest ever, You just did all the comparations anyone would ever need, You´re the best Steam Deck channel
Thank you for the kind words. I'm just happy that I can help 🙂
Awesome video! This may be mentioned in another comment but for the fan volume measurements using a different scale for the dB might give a better visual representation of the perceived loudness. Especially for people not used to sound pressure level measurements or nonlinear measurements.
It seems I can't win 😅 In the last video where I measured dB I got a ton of complaints saying it should be dBA for more accuracy to the human ear. I got a meter capable of reading that and used it here for that reason 😂
@@cryobyte33 Sorry for the confusion the measurement technique is great. There are just conversions from db to sound intensity or perceived loudness that you can use. But you are right you can't please everyone haha either way thanks for the great content.
@@chasemcardle Yeah I don't mind swapping, it's just hard to please everyone without infinite time and/or data 😅
This is so cool, thank you for having the JSAUX. I have both, although I noticed that JSAUX might have taken into account one of my emails to them, because the RGB backplate (still hasn't arrived, so I can't say for sure), looks like, according to the images on their website, to have a section cut-out on the thermal pad to allow for the original Valve fluid dynamics diagram. If true, it will help a lot. I have also got a cheap (from Amazon) aluminium heatsink (40x40x11) with thermal adhesive to directly put over the aluminium backplate. This setup is perfect to use with the JSAUX also new RGB hub. I know it sounds like an ad lol, but I am just really excited for it to arrive (September according to their website).
Thank you for watching! I love JSAUX products thus far and don't have many complaints. I don't have an RGB plate, but hope to get one after I'm in the US to test.
I'm glad you addressed the potential issue with the vented rear plate. People generally think that more airflow is better, but almost always ignore airflow paths and how stock configurations are designed for the most efficient cooling all round, not just for a single component like the SOC. The SSD and VRMs will almost certainly be suffering worse temperatures with the vent over the fan. If you are able to see VRM and SSD temps, do you think you could do a follow up to confirm/debunk this?
So, I actually mentioned it briefly, but:
1. I tested VRMs, ICs, the SSD and the RAM for temperatures with the vented backplate and saw less than 1 degree difference from the stock backplate
2. A lot of air still gets taken in through the rear vent, even with the vented backplate, so I have very few concerns with device longevity.
It's still a concern, but not nearly as much as I originally thought. I tested the vented backplate first because I was convinced that it would be bad for thermals, but I couldn't have been more incorrect, at least as of all my current findings.
Thank you so much for watching!
@@cryobyte33 Ah I must have missed that part. If you're happy then I'm happy. I have the metal plate version and was thinking about changing it for the new one alongside a front plate swap, but had concerns so I'm glad you did the testing.
Thank you for this video. I have been on a cooling odyssee ever since i bought the deck myself. I tried out all the methods you demostrate in this vid, its really good to have you coming up with pretty much the same results. Cheers mate, keep up the awesomeness.
Edit: With that being said, the weirdest thing: I spent half my time in a desert country, andwhen I'm there the old JSAUX backplate combined with a magnetic fan is what worked out best for me. Not so when I am back in western europe though.
Also, there are concerns with JSAUX new backplate in regards to internal airflow and cooling of other elements, specifically the SSD. I even think to remember that Valve had once mentioned waz back that this was the reason for the absence of a second vent in the stock configuration.
Thanks for watching and for sharing your results!
The locational difference is almost certainly humidity, with drier air you'd need more fan/watt to keep temperatures down.
As for the new backplate:
1. I tested VRMs, ICs, the SSD and the RAM for temperatures with the vented backplate and saw less than 1 degree difference from the stock backplate.
2. A lot of air still gets taken in through the rear vent, even with the vented backplate, so I have very few concerns with device longevity.
Hopefully this helps 😁
that was great man, you put a lot of work into this. instead of filling it with fluff and bullshit like most videos.
Excited for the new content as always man!
Hey man, welcome back! Hopefully you like it!
I absolutely love your content man. No other content creator I know of does such extensive and useful testing as you do for the steam deck. Keep up the great work as always, and I wish you the best going forward 😁.
Finally the new jsaux backplate reviews are coming out. Thanks so much for your detailed videos. I'm testing out a devaso backplate that has a metal plate attachment for a thermoelectric fan. Similiar to the jsaux magnetic fan but bigger
I hadn't heard of that backplate, I'll try to check it out! Thank you for watching and I hope the results help you out 🙂
I’m curious as to how the KingPin paste would perform. Very thorough and informative video, thank you! I’ll be keeping the stock backplate but I think I’ll move up to the thermal pad 👍🏼
Dude as soon as you said 50 hours I had to comment, thank you, in advance!
Great video, watched the entire thing, I do have a comment about the backplate though. I'd be interested to know if it could be resealed, how it would perform then in comparison to the stock plate. The negative pressure you point out, according to Gamer's Nexus, that allows area to be pulled over the battery circuit which if it doesn't do that, may be long term implications and could cause damage due to running at higher, possibly out of spec temperatures. If the newer backplate's, fan vent, was sealed, you may get the best of both worlds, and address the dust and noise concerns. Worth considering for sure - looking forward to the block testing :)
While dropping ~5C is definitely awesome. This just goes to show you how well designed the Steam Deck is from the get go. Valve really has some amazing engineers on this product.
Epic vid! I was interested in getting one of those new backplates, and this convinced me fully. Now I’m also considering getting the thermal pads too!
Thank you for watching and commenting! I'm happy that I could answer a few questions for you and hope that you enjoy them as much as I have!
I was thinking about getting the PTM7950 and new JSAUX backplate, but was concerned what the actual games were. This video definitely helps make my decision easier. I might just get the backplate and use some thermal paste I already have. Don’t want to setup the PTM7950 if it is all going to come out for a p3r.
FYI I saw Salocin900 posted a video of an updated Steam Deck which has different mounting screw locations for the heat pipe. Might be an issue for the p3r if Valve has significantly changed the design.
whats a p3r? Would like to know what you mean as im thinking of ordering the thermal pads but have some mx6 paste here? Cheers
Thanks a lot for your extensive testing and the detailed video, earned my subscription :)
I got a few questions if you don't mind:
1. About the airflow from the original vent to the other components (VRMs, ICs, SSD, RAM), even though in your testing there was little variation in their temperature with the Jsaux vented backplate, there's definitely less airflow there, so would you say the additional airflow pull from the Jsaux clip-on fan compensates for that? I'm concerned about long term component longevity with less airflow. If not, would just closing that additional vent with tape make it like the stock backplate or is there more to it?
2. In the new vented backplate, the metal plate is not exposed like the old one. Is there nevertheless a significant advantage attaching a magnetic fan to to it instead of the metal plate directly?
3. Without overclocking, will I get significant performance advantages by thermal modding the Steam Deck?
Thank you!
Thank you so much for watching and subscribing!
1. The clip-on fan SHOULD be able to compensate, especially since it would cover the vented hole, but the tape trick would work too. That said, I've been using mine OC'd with the JSAUX vented backplate for 3 months now with no issues!
2. The new backplate has an aluminum plate rather than steel, so it won't hold the magnetic fan. It also has much less thermal mass and a smaller thermal pad, so the magnetic fan would have a much smaller effect than the old plate.
3. You won't likely see performance advantages without overclocking, but you would have much quieter fans and marginally better battery life as a result!
Hopefully this helps, thank you again for watching!
Seeing you’re video hit my sub box always puts the biggest smile on my face because i know how much hard work and time it takes testing things like this take, And someone with OCD&HDHD I really appreciate the effort you put into your testing with all the different independent variables there is when it comes to testing the smallest of things. I’m planning on water cooling my steam deck in an upcoming video if you ever have the free time I have some questions about attending to dd this stupid idea of mine
Thank you so much for watching the videos and being subscribed! I'm happy that you enjoy them as much as you do, and that it can hold your attention since both my wife and I have ADHD.
As for water cooling, feel free to either email me at my business email, or join the Discord server and start a thread to chat through it 😁
I've got the old JSAUX backplate.
Going to get the recommended thermal pad and new backplate. Thanks so much!
Are there any updates on your concerns with cooling? I have the backplate myself and have had it about 3 months now. No issues on my end but i am by no means running testing on it.
Nothing thus far, it's still my daily driver with no signs of issues!
@@cryobyte33 Thanks for responding! I was gonna ask the same question to see if your SD spontaneously combusted or something. Based on responses from Reddit, they make it seem like all the other components will die due to increased heat. They love saying how valve did negative pressure for a reason etc.
@@cryobyte33 thank you so much for replying. Your work is greatly appreciated. Huge fan here.
I also have had no issues and the lower temps have been awesome.