When it comes to gaming it's not about mega transfers. It's about latency, and less mt can run at lower latency and power, leaving more room for core to power up. What can one do for those who don't know
The funny thing is, most TH-camrs grab a 4090 and pair it with out of date parts. Ontop of that they leave everything stock. that only helps the consumer on a base lvl. not on a advanced lvl. but to be fair, your not an advanced user until you put in 4-5 years of research & the physical work.
But thats not Jays fault, thats the bullshit from Nvidia with their stupid new Powerplug, everything was fine before but hey why WE dont send 600w of Power through little tiny Pins 😂
I met you at CES this year. I told you that I enjoyed your channel and appreciated your work. You asked me what I like about your channel. This video and others like it is what I like about your channel. Thank you.
What do you like about this one??? Him making fun of someone that is going through things lately and maybe his content quality dropped a bit but still tries?? Or that he was talking crap about Asus but then he cosied up when he met them and chickened to open his gob and let it all out. Because I am 101% sure he would chicken laughing at Jay's face about port royal too.
@@selmevias1383 Jay def has integrity. He cut a sponsor out of his channel. Some little boy here trashed them on the Dark Hero "review" and then cosied up on them... mind you the sponsor of the first sentence is the same one this one cosied up to. I just hope this video has nothing to do with it. (/insert conspiracy level:medium)
@@selmevias1383 well... you seem to be arrogant about a few things. By the way... I hit over 8000 stable for over a day until I stopped the test myself on ddr5 on a Z790 Formula with an SP95 14900K (MC83). I advise you to watch some of Jay's vids and support the dude. At least he is not trying to sell you consultation over building a pc for the price of a new cpu 😅
Look, I am NOT a 1%er, an Esports Gamer, nor will I spend $6K or $10k on a computer, just being up front with you. But I AM an engineer, I want the best for my money, and I like HONEST and ACCURATE information. My builds buy your standards are definitively sub-par (currently an Asrock Nova with a 14900K and slower memory) but they are built with information I hope is accurate and they are the best I can get for my money. I appreciate honesty and candor, and would love more videos tearing apart any major tech-tuber for in accurate information as Steve @GN, Linus etc. Thanks for you work and I am sure your community will grow! :-)
*I NEVER use synthetic benchmarks to determine the gains or losses in hardware. I use MSi Afterburner with OSD and run the exact same environment to get an average. Ie: GTA 5 and driving at high speeds on the highway through the city on the same route. Where you typically lose 25% framerate.*
Also, my 2x16 GB (32GB kit) of G.Skill Z5 Trident 7,200 ram was cheap. $130 USD. I wish the tRAS was lower. But the 115 I have is good. For the Corsair Dominator Platinum with the EXACT same specs has a tRAS of 96. But for $100 more. No thanks.
I just bought 7200 memory for my 7800x3d that was on my MB QVL list. I ran it stock(4800) to see results. Then I turned DOCP even with non tuned memory I see pretty significant gain. My system overall feels way for responsive. Thanks for the video you got a new subscriber
Basic common PC knowledge... except there's basically zero real world difference between 7200 and 6000 on Ryzen. Should have saved yourself some money.
"If you don't know what you're doing, you shouldn't own a PC." That's like saying if you don't know how to rebuild your engine, tune your own A/F, add or take timing out then you shouldn't own a car. What a stupid thing to say. It's no different. Install your RAM and enable XMP. Buy a car, turn key and drive.
As a pc builder my self the only thing i wanna say is THANK YOU BROOOO! You have no idea how many pcs built by people who have no idea on the true needs in gaming and productivity are coming to me to get tuned or fixed.
He also used AI CPU overclocking which I can almost guarantee would not be consistent swapping from one ram kit to another. I don’t know how tangible of a difference that would make but in principle, it is definitely a variable that needs to be avoided when you’re conducting a test with intent to prove something.
but thats the most likely thing that a good amount of users will do. most people dont have much experience and will do the basics to get performance boost. the way i see it jay was doing the testing from a new user pov and it wasn't as mad as everyone is making it seem. he should have clarified it but overall not that bad.
@@tomgreene5388 Overclocking is not the issue, it’s the Asus AI method that is the issue. The AI is using the hardware at its disposal to decide what it can get away with. When you change ram kits this drastically, you are changing the conditions that the AI is using to determine its best case overclock. Now, if we were using a static overclock with static voltage, that would be more of a constant for the test. Again, no matter how big or small of a variable it is, it is still a variable that needs to be avoided when you’re trying to make a point exclusively about ram speeds.
@@tomgreene5388 He could have protected himself from a lot of criticism by stating what you just stated, when you don't mention it's for neebz only don't act all bothered when you get memed on for it.
Hi Framechaser, I wanted to know if you could talk about the premature degradation problem of the 13th and 14th gen i9s Intel even changed the specs on the official page for example the 13900ks went from 5600mhz all core to 5400mhz! Have you noticed similar problems? Thank you in advance for your answer.
I had the same issue with my 13900K. The CPU was completely unusable. Even installing C++ or Steam were causing hard lock or even BSOD. So the CPU was RMA'ed by Intel. Since then (October 2023), I have a 13900KS and it begins to have the same behavior ... Changing SVID Behavior or LLC does nothing. Raptor Lake generation is as fragile as porcelain, unbelievable that there is no statement by Intel so far. And yet Reddit or Overclock net posts appear constantly on this subject... I've got a multitude of chips of the Skylake Era and no one failed like that even overclocked with high daily voltages ... Using a Raptor Lake SKU during 6-9 months without any OC and the CPU is wrecked ... What the hell Intel ...
Been running multiple 13900KS and 14900K chips at 5.7-5.8 all core for a long time and have no issues with crashing. Crashing is likely because cooling is simply insufficient.
It’s actually a blessing watching your content for the facts because I’ve probably build like 15 pc’s in my life experienced a lot of bs and what I’ve concluded from every expansive life lesson is that ryzen gets gassed up for being the best but in reality intel is always better and most of all way more reliable. Almost everywhere you look people will tell you ram don’t matter when in facts it matters a lot if your going for high fps. And lastly amd will always get overhyped because Nvidia is to expansive but they are also simply better mostly because of reliability. It’s awesome to watch someone telling the actual facts I’ve experienced everything first hand and sadly found your channel afterwards but if someone asks me who to watch for reliable source it’s gotta be jufes. Thanks for the great content and keep it up!
Uh hold up Cees, I tel used to be the best. But as far as current ryzen stats and watt per $$ AMD is definitely mopping the floor with 7800x3d compared to 13th gens etc. Intel pushes 250+w whole AMD under 100w all day long on similar bench tests
@@Giggidygiggidy12 you give the answer yourself amd is the best performance to dollar whise, amd is also the best performance to watt ratio, but if your talking simply what’s the best processor money can buy it’s still intel 7800x3d can compete with intel when you talk about optimized games yes but a 10th gen or 14th gen intel will give you a few fps more in an optimized scenario so i will always look at games wich are highly cpu demanding and then intel will always win they don’t have the am-dip and they can push ram way better wich also impacts most cpu demanding games. Also the way intel simply works and amd is always fiddling around turning this ccd off to get the game to run on the other ccd is just annoying they don’t have it windows optimized at all. And then you buy an expo rated ram kit nothing crazy you set expo on and just want to play games it will run fine for 10 starts and the 11th one it will crash for no reason and your stuck in memory training for 10 minutes.
I love Jay, but he makes some really weird and stupid mistakes. He made a fan "guide" talking about the types of fan bearings and which are best, but confused fluid dynamic bearings with sleeve bearings. He was actually telling people to stay away from fans like Be Quiet! and Noctua with FDB bearings and to go and buy cheap sleeve bearing fans! It was SOOOO bad. Anyone with grinding fans in a few years who watched his video are going to be pissed. Pretty sure my fact check comment was deleted or buried. XD
I just found your channel hilarious reaction. BTW I remember a couple years ago watching tech tubers and yelling at my screen "you fing idiots!! What is wrong with you??" Now people are finally calling it out. The Verge all over again.
Wow. "Cant make this shit up." That statement says it all. I saw the video and thought he was on crack. Why would companies sell faster ram? Faster ram exists for a purpose.
You're the only PC youtuber I still watch since I got into building pcs in 2013, you still seem to have the passion for high fps cpu limit breaking pcs. I find that most LinusTT-like tech youtubers are a plague on this site as they always feel the need to add 7-8 minutes of pointless context or ad reads and a lot of them don't really play games much anymore and they tend to focus on other areas of their life. If they'd only make videos less than 5 minutes and only talk about issues they had or why they tested a certain way then show the dang fps numbers then maybe I'd watch them again. Otherwise I'd rather spend the time myself to crunch the numbers and find what works. Even the 'good' channels like gamers nexus think it's ok to not even use 7200MHz ram when testing the i9 14900K against AMD, just because AMD can't easily hit those numbers doesn't mean they shouldn't use better sticks for intel. Keep up the good work!
LOL!!!!!!! I knew someone would end up calling out Jay on that video, and alas the youtube algorithm recommended a video from a channel that did just that. That was so hard to watch I was laughing. As soon as I saw the 6400 being beaten by a JEDEC setup I knew something was amiss. I assumed it was just a junk kit of RAM with high speed but atrocious timings, but ya know, you're probably right it was just unstable and he didn't bother to stability test it before running benchmarks. 8:32 lol, come on man that's not the AMD experience. Unless you buy a shit RAM kit or have a good one but don't enable EXPO, then yeh, it's brutal but then 4800 is just brutal on both platforms. So brutal you can FEEL it like you showed on your video. A proper 6000-6400 kit with proper EXPO timings is a great experience however.
he literally said he wasnt talking about the hardcore esports players.. but the podcast i watched with you in it the other day even the guy said you tailor to people trying to spend 10,000 dollars on a single setup while other tubers try and save money for people.. he was spot on.. cause what i got from your video just now is spend 150 on ram while jay was saying pay 80 dollars and get a better cpu
I love it when these tech tubers do RAM tests with no knowledge. Its always synethatic benchmarks and GPU Bound scenearios like that's going to show anything lmao.
Legit just commented this on someone else’s comment lol. They are the first to pick a 4090 & pair it with shitty parts & not even tune the system. and that only helps on a base level, not on a advanced lvl. but to be fair if you haven’t put in 4-5 years in this field then your not an advanced user. but we have people like jay so it doesn’t even matter 😂. clearly years doesn’t matter
how often are you cpu limited playing games? yeah they could test it in a way to show the actual differences in the ram but if its not going to matter in 95% of scenarios then who cares. he showed that for most people the ram won't matter much and for the people that it will make a difference, well they are already know that. you people are so quick to try and point out a flawed testing method that you don't even realize that the ting you want is just as far fetched of a testing method. from a new or beginner perspective his testing was completely fine.
@@tomgreene5388 that’s the thing, a lot of people are not beginners. Hense why he’s roasting jay. He most likely feels like he should know what he knows 🤷🏼
@@tomgreene5388 there’s people out there that want a rock solid 144 fps experience out there with no stutters or frame drops. A lot of modern games these days that struggle to hit 144 FPS such as Starfield, Hogwarts Legacy, Star Wars, Battlefield, Need for speed and so on. A good memory kit is only a bit more expensive but can make your 1% and 0.1% frame times so much smoother and massively reduce stutters that occur mid action or texture streaming.
@@silfrido1768 if yall think jay doesn't know the info you are stating then you are trippin. of course he knows it, he's done all levels of overclocking and benchmarking. i just know that most of his videos and info is shown in a new user friendly way. he for sure should have at least said that there is ways of testing to see the differences but i don't think its as bad as people are saying. and this is coming from a pretty nuetral viewpoint, i don't even really care for his videos all that much.
Jay isn't clueless, he must know better. At bare minimum, faster RAM will improve your minimum and low frame rates. Which means it's actually one of the most important components to have configured correctly for smooth game play.
Damn I actually found a youtuber who know what they are talking about and they game !!! Sorry at Frame Chasers but I had to smash the subscribed button and looking forward to future videos
I wanted to mention this in the comments of Jays video when it came out but got distracted and now it's 2 weeks later and no longer relevant, but the performance uplift from faster memory is platform dependent. Yes, there isn't much benefit on Intel or Ryzen 3D CPUs but on Ryzen non-3D chips, you should match your infinity fabric speeds and run the best latency you can afford because this can be a difference of up to 15% in games. But this ignores the fact that most people still aren't stressing their CPUs in a meaningful way because the GPU or the display are almost always the bottleneck.
LMFAO... When I saw his video my first reaction thought "Buildzoid would go CRAZY if he watches this shit" LOL Funny thing is can't find any of the comments saying the same thing, might be deleting them 'cause he's very emotional.
look at 7:47 right under where it says Memory Type is your clock speed and it says 2400 MHz not 4800 MHz which it should for your test to be correct and then CL numbers should be 40-40-40-72
Mister, computer game players are not just those who call themselves competitive players. I've been playing for around 30 years and for me what matters is the highest possible resolution and the best image quality that can be obtained at a minimally acceptable FPS number, which varies depending on the game in question. I've been playing at 4K since 2016 and the graphics card is always the bottleneck. That's why I only spend the minimum amount on the CPU and its memories so that these two components don't become a major bottleneck at 4K.The money I save with this option all goes towards the graphics card! And Jay, around 15 minutes in, talks about you!
Yup and the video about “Does cleaning your old GPU improve performance” was even worse. They observed high temperatures and after taking it apart it was then disclosed that one of his friends attempted to service it once before and screwed everything up. Missing thermal pads, loose screws, who knows what else. Video should have been about a sealed never opened GPU from say 2017 to see exactly what things look like inside, not some mess that someone else already did once before. Jay does have health problems so maybe that’s why such low effort content but the RAM video blew my mind how bad it was.
you're using a 4090? of course you'll have better performance. the Gain you get from ram high ram speeds for a someone who is NEW to PC gaming is negligible. 10-15 fps for $100 more is not worth the price especially if youre new to pc why not spend the money elsewhere; that's the point of jays video. I respect your overclocking videos, you don't have to click bait for views
He is no different from LinusTechTips, all they care about is money & views. I'll go one step further Jaytwocents channel is not even doing good anymore he has 4million subs and is getting 100-200k views on a video that is terrible not even a quarter of the people clicking on his videos.
I generally can't really handle your videos because of the way you rant. But for once, I would say that this video is quite good with minimal ranting and more to the point which I like. Please make more videos like this.
I think the biggest thing that is getting missed here is that there is a spot in the middle that probably makes more sense. 6000MHZ RAM with good timings can be purchased for a decent amount less than 7200+. The money saved could be put to upgrading your CPU/GPU, or some other upgrade that you would feel more than having the fastest RAM out there. For a comparison I looked at some fairly standard G.SKILL 32GB/64GB Kits in both speeds and there is about a $40/$70 difference. In budget builds the extra money can go a long way... in systems like yours/mine with everything else maxed out... it might show more of a numbers difference, but can we really see the difference of anything above 280 FPS? That being said i have a a $300+ Corsair Platinum RAM kit... so not sure I should be talking.
I saw Jayz thumbnail and title, chuckled to myself and thought I can't wait until Jufes sees this one lol Thanks to you Jufes I knew not to even bother clicking the video. 👍
Mate, it’s ridiculous how these guys get so big with the information they throw around. Good to see you calling it out dude! Keep it up pal! Big support from the UK.
It's even worse with stock market influencers. Even legit guys worth $300 million begged their audience not to buy Nvidia stock at $400/share and today it's $800/share. I know it's unrelated, but if you can't think for yourself in 2024 you're gonna get wrecked in all your pursuits.
@@ryoannazuk949 ohhhhh ok you’re just a super fun troll. what a silly cutie. all you wee little trolls know Nvidia is the third largest company in the world 😘
@@ryoannazuk949 your diction and syntax and reiteration of common Reddit platitudes convey your ignorance, how many shares do you own? What was forward p/e when you purchased? Have you purchased put protection to hedge your investment? What impressed you most about the the last earnings call? The only meme here is your adorable trolly sentiment. Best of luck!
farcry 6 - tomb raider - bf 2042 - warzone and the list goes on from jedec 4800 to 6800 - 7200 with a properly tuned tier 1/2/3 timming , it's a 15 to 35% more perf , plus you get better latency overall from 70- 80ns to 49-55ns , that's why i loved my 12900k delided to 5.30 /4.20e /4.4 ring , i'm rocking the 220 300 fps on both warzone and bf2042 plus a snappy windows experience .... . fun fact an oc12900k 5.40 with a properly tuned rams 6800 -7200 beats the 13900k + 6000 mhz rams stock in tomb raider :D
And the worst part i think Jay believes he is right with this memory speed thing, i don't even know how his video was sponsored by a ram manufacturer and that manufacturer hasn't got mad because Jay dished their fast memory.
In modern warfare my 7800x3d gained over 100 fps from tuning my fclk, bclk, llc on low-medium, overclocked my 7200mhz cl 34 to 7400mhz, - pbo offset of 25 & a few other settings in bios
Thanks! As much as I do not enjoy bashing on a particular person. I cannot deny the evidence and the testing you literally did live. I appreciate the edits on youtube and all the information. Everyone, no matter how shady they are will provide good information but eventually the foundation cracks. Money/click-bait becomes the drive for many people.
I absolutely love your no bullshit channel this means allot to have someone like you doing content like this that we as consumers don't get screwed over. I would really like to do a paid session to help me tune my system but maybe at a later stage as I also got screw over financially but will definitely do this at a later stage this year. Thank you for your content buddy you are awesome.
Let's not forget that Jay was the MAIN fear mongerer that brought forth the entire "fear the POSCap" situation with the release of the RTX 3k series even though POSCaps are higher quality..
@@bigpoppa1234 If you're talking about the video that waas released 3 years ago titled "The Real Problem with RTX 3080 Crashes is NVIDIA'S Paranoia of Leaks" wherein at 0:40 he states: "There's actually been more development to this story, so the addon that you're not going to hear me talking about in the biking video witht he capacitors and crashing stuff; is that Nvidia's released a driver now, which according EVGA, their six cap cards are no longer crashing with that driver change by Nvidia when they were before, so, there's more to this than just the caps"; and later at 4:12 he states "basics of the issue: there's one issue we started researching right before we left, we only had an hour researching it, it's not much time, where the partners are pushing frequency too aggressively like over 2Ghz, and cards will crash when they peak. That's one issue, other one, is the capacitors on the back of the GPU socket being problematic, and I'm not going to talk about the electrical side of it. I'll leave that to other people, but I can talk about like, why this happened, and it's a bit of a partner's fault, a bit Nvidia's fault, and of course partners are very fast to throw Nvidia under the bus, Nvidia, I'm sure will throw partners under the bus..." "As for the part that's nvidia's fault, a couple things. One of them is that Nvidia's so paranoid about leaks that it's hurting itself", etc. I'm afraid you misunderstood the point of their video entirely. Steve wasn't saying that the poscaps were the issue, he was saying that their overly aggressive use of them as a filter was a PART of the problem paired with the fact that Nvidia was so paranoid of leaks that they didn't give AIB partners enough time hands on with the drivers and manufacturing specs to PROPERLY design the cards, and technically speaking he was right on the money. With the RTX 3k release Nvidia told AIB partners that "In order to sell a card of this series you MUST design a new heatsink that cost relative amount in R&D and production as the one we've manufactured" then gave them the information they needed to do it SIX MONTHS later than they normally do giving them an entire MONTH - just one - to figure EVERYTHING they needed to out about these cards, and as you can imagine, that's simply NOT enough time. Adding to this now farced inflation thanks to Ngreedia we had companies SCRAMBLING to try and push out an AIB model in order to try and even stay RELEVANT; and the result was a lot of cards that simply weren't designed properly in a number of ways with one of them being the poscaps. Don't get me wrong, I'm not defending Steve, or his stance on the issue, but if I'm being frank he has/had a MUCH more quantitative understanding of the issue than most any other "techtuber" that was covering the situation. Lest we forget, Stephen Burke is one of the only ones of them that actually HAS a background in electrical engineering. The rest of them are pretty much just "self-taught plebs" and I'm in NO way bashing "self-taught" because I'm one of them as well, AND an electrical engineer; but I will say that a proper HISTORY of working with electrical applications and theory helps a hell of a lot more in understanding how this issue came to be than a microsoft certification for software obtained during your tenure as IT. Forgive any typos, I tried not to spend too long typing this up.
I’ve been eagerly waiting for this video from you. When I watched the first 5 minutes of Jay’s video, I knew that you would have a future roast incoming 😂
My favorite thing is when people are whining about FPS while the CPU and GPU are under 80% usage, and when you say something about memory bandwidth, latency, XMP/tuning, etc, you get the internet equivalent of blank stares and drool.
I need help please… I want to build my first gaming PC. I play Skyrim mainly with a shit ton of mods. I want to be able to play in 4K and use the new AI mods that are coming out… I also want to be able to play other games I’m looking at. I will need this to also be used as a home/home business computer. I’ve seen many say to wait as new and much better components are coming out soon. Should I build now or wait? I want to build an all White system. Any help would be very appreciated. Thanks and have a great day.
@FrameChasers wouldn’t playing at 1080p have more impact on the cpu vs playing at 4k? Would that not tax the cpu ram a bit differently than playing at 4k and having the vram on the gpu used more? I seriously don’t know and would be interested in watching or hearing more. Just trying to learn
this video only show 1 game.. while Jay show multiple benchmark..
8 หลายเดือนก่อน
I'm sorry, but my TEAMGROUPs on 4800Mhz do not cause such a problem as yours. And I have the resolution set to 4K. Personally, for example, I see almost no difference in 4K between 6000Mhz and 7200Mhz. Maybe a lower FPS drop in some places. I tested BF2042, Forta Motorsport, Forza Horizon 5, Cyberpunk 2077. The only thing I noticed is a lot of stability work at 7600Mhz! Anyone who has a board with 4 DIMMs and RAM 7200 Mhz+ has rather invested in the future. And yes I have an i9 13900k and an RTX 4090.
How much difference does it make in 1440p ultrawide (high/ultra)? I know most of these benchmarks are done at 1080p Low/Med to purposely induce a bottleneck but I haven't used 1080p in a decade and I never use Low settings ever. I just care about my real world results.
@@stevesonger1Did you manually tune your timings and voltages, that’s what Jufes is showing here. Speed+Timings+voltages Tuned= potentially ~25% gains.
Yea he definitely doesn't play games. Anyone who says how unlikely it is to get CPU bound at a certain resolution doesn't play games. But that is still nothing compared to getting a result where 4800 is faster than 7200 and just running with it.
@JesusisLordSEEK What cpu/gpu are you running? I have a 7800x3d with cl30 6000mhz ram and was debating on messing with the ram a bit but idk if it's really worth it.
Hold on. Didn't Jay's chart shows that faster ram gives more fps at 1080p? Your test was also at 1080p. So what is the problem? not shelling for anyone but am I missing the point here?
I've seen 2 or 3 Jays videos, after last one I said to myself no more of that guy... ppl can't be this stupid right? now I just found this, and hope in humanity is slowly recovering. Btw, Im a total noob in PCs, and total pro in humans. Thx for this one! cheers
I feel like jay started a while back making his employees make the videos, and he just comes in to do a 10 min talk about the graphs they produced while not giving a shit if its true or any good data. He simply doesnt care about the videos anymore and it shows straight through. He made some cool content back in the day but its a long time since I was ablo to watch any of the recent videos released. I dont recall him even mentioning cas latency, only speed. Thats just insane from a ram video
I'm the only overclocker I know of that actually plays games. I've never got the best results from XMP. It's usually better than stock motherboard settings that go low speed but I usually speed up the cache which you need to do to really take advantage of the faster ram speeds. Tight timings are normally unreliable for gaming. I usually enter the XMP timings manually on the primary's as the board will usually set better secondary timings than XMP. You need balance for the optimum system performance and reliability. Buildzoid is awesome for getting top RAM performance but his reliability sucks for gaming. Go too fast on the memory and the controller will get hot limiting your overclocking capability on the cache and cores. Even if a CPU doesn't throttle excess heat can still hurt performance. It's the price of fast ram that is usually the limiting factor, but generally faster ram will net gains and usually raise the 1% lows for smoothness.
Amd CPU work better better with a increase ram frequency so you will notice bigger performance differences on that platform because amd cpus utilises the speed increases better. So you are more likely to notice a performance difference on a amd platform going from 7000 to 4800. Intel vs Intel platform you might get 1 to 25% differences between each platform. So you won't notice much performance differences.
Yeah, it's pretty bad, but to be fair, there's a _violent_ difference in his audience and your audience. The average normie that watches Jay or Linus isn't tuning their 4090 and direct die cooled 14900ks to get 500 FPS in CoD at 1080p. They're buying a 4060 and a 12600 on a cheap b660 to play Hogwarts and Palworld at 1440p ultra. For that customer they *_won't_* see a real difference in DDR4 and DDR5, let alone 4800 and 7200 DDR5 and it would be stupid to spend twice as much on your memory when it's far better spent towards the GPU. I build a ton of systems and 99% of them are on a hard budget below $1500 or lower in these times thanks to Bidenomics. For the overwhelming majority of users that can't afford even a 4070ti I have a huge difference in what I recommend and what I buy for myself. It's easy to go all PCMR and say "don't be poor", but I will do anything in my power to get another gamer off of Gaystation and into PC gaming.
Eh even ltt showed that faster ram among ddr5 gets you appreciable fps difference. That's the video where I settled on a cl30 6000 for the best performance to cost for me. The real flaw with techtubers for normal people is they err on the side of caution since their audience doesn't have 3k to spend on the best of the best components. I even take that into consideration when I use their information.
I don't get it. If I play low quality 1080p on 240Hz+ in CS, yeah the bottleneck is there where the ram makes the diff, but if I play a resolution and game that actually is GPU bound, that's where it's a waste of money... Now I'd say both Jay should be more clear as well as your video... Some disclaimer...
He's right. Why spend more money on faster memory if its not needed? that fps boost from 279 to 350 makes no difference to you whatsoever. Maximize the memory you have, if you need to before bying a faster one. But if you are already in 300 FPS mark, to most gamers it wont make a lick of difference.
As soon as I saw the title of Jay's video I knew shit was going down xD Glad I followed you that other time you called out Boxed on the memory rank debacle.
Dufus I was laughing so hard when you lowered the ram to 4800 and you get the same windows experience as I do EVERY DAY ARRRRR with the 7800X3D with memory at 6400. So wished I found your channel before upgrading from Zen 3 to Zen 4.
I want to upgrade to DDR5 now but I also need a motherboard. I was thinking of getting: CMP32GX5M2X7200C34 7200MHz CL34. What is one of the most high speed RAM stable motherboards out there? I have a DDR4 system now. Other specs of my system: 13900K, Arctic Liquid Freezer II 420, 4090, MSI - MEG Ai1300P PCIE 5.0 PSU. I want to upgrade because I'm still on ASUS TUF GAMING Z690-PLUS D4 motherboard and DDR4 3600 14-15-15-35 TF10D432G3600HC14CDC01. Looks like it's worth it upgrading to DDR5 but Intel might release a new socket soon.
if you're planning on buying a new motherboard to upgrade just wait for the next socket update and go from there then sell your current mobo CPU and Ram as a package
Jay and linus no longer techtubers theyre more like journalists/entertainers nowadays. I've got to say, you hurt my feelings all the time, cuz i have an all amd rig but I Love your content man, its pretty based
Ok, as someone who has been building his own PCs since 2010, i feel really worried, because i don't know what was wrong in Jay's video. I watched it like yesterday and now i got this video in my recommended. I'm not a tech expert or anything. When i choose my PC parts, i just go for the basics. The more the better. Be it MHz, GB, in anything. Seems like i got a lot to learn still...
His numbers didn't add up. He's espousing that more speed on DDR5 at the ones he tested yield no gains, but there's an obvious issue, as anyone who has validated the numbers does not see these results. Tightening timings and speed bumps typically show clear scaling. In Jay's it did not which means something was wrong with the set up. DDR5 is notoriously tough to tweak, and that's pretty much all that's illustrated in Jayz video.
Its amazing to me that these folks wont consider GPU-bound processes, that use VRAM, will see the least improvements. I mean, increasing system RAM has no effect on the VRAM - i would have hoped these youtubers would be more honest about this
GN tries to target a more mainstream audience. Your audience doesn't really translate through to any mainstream. Maximum fps really isn't mainstream, since it's so expensive. Spending 4k$ on a pc every 2 years isn't in 99.9% of peoples budgets.
![ไข้ขึ้นเลย! เมื่อแฝดโดนจัดเต็มวัคซีนครบ2เดือน🤒 [cc] แดนแพทตี้ SS2 | EP.54 |](http://i.ytimg.com/vi/z2VHItnyyvY/mqdefault.jpg)
When it comes to gaming it's not about mega transfers. It's about latency, and less mt can run at lower latency and power, leaving more room for core to power up. What can one do for those who don't know
Bro Linus and Jay complanin' 'bout nvidia's charts and then doin' the same thang is ironic af
Exactly
The funny thing is, most TH-camrs grab a 4090 and pair it with out of date parts. Ontop of that they leave everything stock.
that only helps the consumer on a base lvl. not on a advanced lvl.
but to be fair, your not an advanced user until you put in 4-5 years of research & the physical work.
Preach@@silfrido1768
great point lol
Yeah, these two are definitely entertainment channels lol
Jay: "You NEED this Cablemod 90 degree adapter."
Cablemod: "That was expensive."
But thats not Jays fault, thats the bullshit from Nvidia with their stupid new Powerplug, everything was fine before but hey why WE dont send 600w of Power through little tiny Pins 😂
I met you at CES this year. I told you that I enjoyed your channel and appreciated your work. You asked me what I like about your channel. This video and others like it is what I like about your channel. Thank you.
What do you like about this one??? Him making fun of someone that is going through things lately and maybe his content quality dropped a bit but still tries?? Or that he was talking crap about Asus but then he cosied up when he met them and chickened to open his gob and let it all out. Because I am 101% sure he would chicken laughing at Jay's face about port royal too.
@@sv0g391 Integrity suffer no "quality drop" nor "going through things", you have it or you don't.
@@selmevias1383 Jay def has integrity. He cut a sponsor out of his channel. Some little boy here trashed them on the Dark Hero "review" and then cosied up on them... mind you the sponsor of the first sentence is the same one this one cosied up to. I just hope this video has nothing to do with it. (/insert conspiracy level:medium)
@@sv0g391 The video being reviewed here prove that he haven't.
@@selmevias1383 well... you seem to be arrogant about a few things.
By the way... I hit over 8000 stable for over a day until I stopped the test myself on ddr5 on a Z790 Formula with an SP95 14900K (MC83).
I advise you to watch some of Jay's vids and support the dude. At least he is not trying to sell you consultation over building a pc for the price of a new cpu 😅
Look, I am NOT a 1%er, an Esports Gamer, nor will I spend $6K or $10k on a computer, just being up front with you. But I AM an engineer, I want the best for my money, and I like HONEST and ACCURATE information. My builds buy your standards are definitively sub-par (currently an Asrock Nova with a 14900K and slower memory) but they are built with information I hope is accurate and they are the best I can get for my money. I appreciate honesty and candor, and would love more videos tearing apart any major tech-tuber for in accurate information as Steve @GN, Linus etc.
Thanks for you work and I am sure your community will grow! :-)
*I NEVER use synthetic benchmarks to determine the gains or losses in hardware. I use MSi Afterburner with OSD and run the exact same environment to get an average. Ie: GTA 5 and driving at high speeds on the highway through the city on the same route. Where you typically lose 25% framerate.*
Also, my 2x16 GB (32GB kit) of G.Skill Z5 Trident 7,200 ram was cheap. $130 USD.
I wish the tRAS was lower. But the 115 I have is good. For the Corsair Dominator Platinum with the EXACT same specs has a tRAS of 96. But for $100 more. No thanks.
The manufacturers have figured out how to maximize the benchmarks in ways that don't translate to better performance in a game.
I just bought 7200 memory for my 7800x3d that was on my MB QVL list. I ran it stock(4800) to see results. Then I turned DOCP even with non tuned memory I see pretty significant gain. My system overall feels way for responsive. Thanks for the video you got a new subscriber
6000 beats 7200 on ryzen 7000...
Basic common PC knowledge... except there's basically zero real world difference between 7200 and 6000 on Ryzen.
Should have saved yourself some money.
@@guyzer75 Its a CPU with Cache stacked on top of it so RAM latency and speed isn't much of a boost in performance compared to normal CPU.
"If you don't know what you're doing, you shouldn't own a PC." That's like saying if you don't know how to rebuild your engine, tune your own A/F, add or take timing out then you shouldn't own a car. What a stupid thing to say. It's no different. Install your RAM and enable XMP. Buy a car, turn key and drive.
It's more like if you don't know what you're doing, you shouldn't be spreading a bunch of misinformation online for clickbait views
If you don't know how to rebuild an engine you shouldn't be educating others on that topic, no?
Jay is sick. I haven't been watching his videos lately but that may have something to do w the minimal effort.
As a pc builder my self the only thing i wanna say is THANK YOU BROOOO! You have no idea how many pcs built by people who have no idea on the true needs in gaming and productivity are coming to me to get tuned or fixed.
I started watching Linus and Jay. But as i've progressed into tuning my PCs I tend to watch GN, DerBauer, Skatterbench and here most recently.
I mean u only gave useless names for the tuning scene but ok.
Skatterbench hella good cpu overclocker
@@alexandrulascu6436 well maybe share some knowledge instead of just being a troll
add Actually Hardcore Overclocking to that list great content
Damn no mention of buildzoid? I learned the basics of ram tuning from him few years ago.
is the second time crank up the 4090 with afterburner just cut out or do u forget ? have a n1 one
He also used AI CPU overclocking which I can almost guarantee would not be consistent swapping from one ram kit to another. I don’t know how tangible of a difference that would make but in principle, it is definitely a variable that needs to be avoided when you’re conducting a test with intent to prove something.
but thats the most likely thing that a good amount of users will do. most people dont have much experience and will do the basics to get performance boost. the way i see it jay was doing the testing from a new user pov and it wasn't as mad as everyone is making it seem. he should have clarified it but overall not that bad.
@@tomgreene5388 Overclocking is not the issue, it’s the Asus AI method that is the issue. The AI is using the hardware at its disposal to decide what it can get away with. When you change ram kits this drastically, you are changing the conditions that the AI is using to determine its best case overclock. Now, if we were using a static overclock with static voltage, that would be more of a constant for the test. Again, no matter how big or small of a variable it is, it is still a variable that needs to be avoided when you’re trying to make a point exclusively about ram speeds.
@@tomgreene5388 He could have protected himself from a lot of criticism by stating what you just stated, when you don't mention it's for neebz only don't act all bothered when you get memed on for it.
Hi Framechaser, I wanted to know if you could talk about the premature degradation problem of the 13th and 14th gen i9s Intel even changed the specs on the official page for example the 13900ks went from 5600mhz all core to 5400mhz! Have you noticed similar problems? Thank you in advance for your answer.
My friends overclock started crashing after about 6, months is that the. Reason?
@@worldwar208 Maybe what we can read on Reddit, etc... The problem would come from consumption but nothing very clear and no one talks about it!
I had the same issue with my 13900K. The CPU was completely unusable. Even installing C++ or Steam were causing hard lock or even BSOD. So the CPU was RMA'ed by Intel. Since then (October 2023), I have a 13900KS and it begins to have the same behavior ... Changing SVID Behavior or LLC does nothing. Raptor Lake generation is as fragile as porcelain, unbelievable that there is no statement by Intel so far. And yet Reddit or Overclock net posts appear constantly on this subject... I've got a multitude of chips of the Skylake Era and no one failed like that even overclocked with high daily voltages ... Using a Raptor Lake SKU during 6-9 months without any OC and the CPU is wrecked ... What the hell Intel ...
@@ColonelStrikedamn I always assumed Intel cpus were just perfectly reliable. Ive bought used ones and never had a problem without physical damage.
Been running multiple 13900KS and 14900K chips at 5.7-5.8 all core for a long time and have no issues with crashing. Crashing is likely because cooling is simply insufficient.
It’s actually a blessing watching your content for the facts because I’ve probably build like 15 pc’s in my life experienced a lot of bs and what I’ve concluded from every expansive life lesson is that ryzen gets gassed up for being the best but in reality intel is always better and most of all way more reliable. Almost everywhere you look people will tell you ram don’t matter when in facts it matters a lot if your going for high fps. And lastly amd will always get overhyped because Nvidia is to expansive but they are also simply better mostly because of reliability.
It’s awesome to watch someone telling the actual facts I’ve experienced everything first hand and sadly found your channel afterwards but if someone asks me who to watch for reliable source it’s gotta be jufes.
Thanks for the great content and keep it up!
Uh hold up Cees, I tel used to be the best. But as far as current ryzen stats and watt per $$ AMD is definitely mopping the floor with 7800x3d compared to 13th gens etc. Intel pushes 250+w whole AMD under 100w all day long on similar bench tests
@@Giggidygiggidy12 you give the answer yourself amd is the best performance to dollar whise, amd is also the best performance to watt ratio, but if your talking simply what’s the best processor money can buy it’s still intel 7800x3d can compete with intel when you talk about optimized games yes but a 10th gen or 14th gen intel will give you a few fps more in an optimized scenario so i will always look at games wich are highly cpu demanding and then intel will always win they don’t have the am-dip and they can push ram way better wich also impacts most cpu demanding games. Also the way intel simply works and amd is always fiddling around turning this ccd off to get the game to run on the other ccd is just annoying they don’t have it windows optimized at all. And then you buy an expo rated ram kit nothing crazy you set expo on and just want to play games it will run fine for 10 starts and the 11th one it will crash for no reason and your stuck in memory training for 10 minutes.
Jay also said to go out an buy gpus NOW (right before the crash begins).
He is (or was) an EVGA shill.
All of USA cityzen apparently is EVGA shill for some reason. I don't know why people simps for them way too much.
I love Jay, but he makes some really weird and stupid mistakes. He made a fan "guide" talking about the types of fan bearings and which are best, but confused fluid dynamic bearings with sleeve bearings. He was actually telling people to stay away from fans like Be Quiet! and Noctua with FDB bearings and to go and buy cheap sleeve bearing fans! It was SOOOO bad. Anyone with grinding fans in a few years who watched his video are going to be pissed. Pretty sure my fact check comment was deleted or buried. XD
I just found your channel hilarious reaction. BTW I remember a couple years ago watching tech tubers and yelling at my screen "you fing idiots!! What is wrong with you??" Now people are finally calling it out. The Verge all over again.
Wow. "Cant make this shit up." That statement says it all. I saw the video and thought he was on crack. Why would companies sell faster ram? Faster ram exists for a purpose.
You're the only PC youtuber I still watch since I got into building pcs in 2013, you still seem to have the passion for high fps cpu limit breaking pcs. I find that most LinusTT-like tech youtubers are a plague on this site as they always feel the need to add 7-8 minutes of pointless context or ad reads and a lot of them don't really play games much anymore and they tend to focus on other areas of their life. If they'd only make videos less than 5 minutes and only talk about issues they had or why they tested a certain way then show the dang fps numbers then maybe I'd watch them again. Otherwise I'd rather spend the time myself to crunch the numbers and find what works. Even the 'good' channels like gamers nexus think it's ok to not even use 7200MHz ram when testing the i9 14900K against AMD, just because AMD can't easily hit those numbers doesn't mean they shouldn't use better sticks for intel.
Keep up the good work!
LOL!!!!!!! I knew someone would end up calling out Jay on that video, and alas the youtube algorithm recommended a video from a channel that did just that. That was so hard to watch I was laughing. As soon as I saw the 6400 being beaten by a JEDEC setup I knew something was amiss. I assumed it was just a junk kit of RAM with high speed but atrocious timings, but ya know, you're probably right it was just unstable and he didn't bother to stability test it before running benchmarks.
8:32 lol, come on man that's not the AMD experience. Unless you buy a shit RAM kit or have a good one but don't enable EXPO, then yeh, it's brutal but then 4800 is just brutal on both platforms. So brutal you can FEEL it like you showed on your video. A proper 6000-6400 kit with proper EXPO timings is a great experience however.
he literally said he wasnt talking about the hardcore esports players.. but the podcast i watched with you in it the other day even the guy said you tailor to people trying to spend 10,000 dollars on a single setup while other tubers try and save money for people.. he was spot on.. cause what i got from your video just now is spend 150 on ram while jay was saying pay 80 dollars and get a better cpu
I love it when these tech tubers do RAM tests with no knowledge. Its always synethatic benchmarks and GPU Bound scenearios like that's going to show anything lmao.
Legit just commented this on someone else’s comment lol.
They are the first to pick a 4090 & pair it with shitty parts & not even tune the system.
and that only helps on a base level, not on a advanced lvl.
but to be fair if you haven’t put in 4-5 years in this field then your not an advanced user.
but we have people like jay so it doesn’t even matter 😂. clearly years doesn’t matter
how often are you cpu limited playing games? yeah they could test it in a way to show the actual differences in the ram but if its not going to matter in 95% of scenarios then who cares. he showed that for most people the ram won't matter much and for the people that it will make a difference, well they are already know that. you people are so quick to try and point out a flawed testing method that you don't even realize that the ting you want is just as far fetched of a testing method. from a new or beginner perspective his testing was completely fine.
@@tomgreene5388 that’s the thing, a lot of people are not beginners. Hense why he’s roasting jay. He most likely feels like he should know what he knows 🤷🏼
@@tomgreene5388 there’s people out there that want a rock solid 144 fps experience out there with no stutters or frame drops. A lot of modern games these days that struggle to hit 144 FPS such as Starfield, Hogwarts Legacy, Star Wars, Battlefield, Need for speed and so on. A good memory kit is only a bit more expensive but can make your 1% and 0.1% frame times so much smoother and massively reduce stutters that occur mid action or texture streaming.
@@silfrido1768 if yall think jay doesn't know the info you are stating then you are trippin. of course he knows it, he's done all levels of overclocking and benchmarking. i just know that most of his videos and info is shown in a new user friendly way. he for sure should have at least said that there is ways of testing to see the differences but i don't think its as bad as people are saying. and this is coming from a pretty nuetral viewpoint, i don't even really care for his videos all that much.
Jay isn't clueless, he must know better. At bare minimum, faster RAM will improve your minimum and low frame rates. Which means it's actually one of the most important components to have configured correctly for smooth game play.
overclocking an XMP 5200 to 6400 / 6800? possible? Corsair? Don't want $hitstone in my pc ever.
Where is the Asus Ai overclock vs manual OC Video??? Would love to see that. Good work so far.
No time
Damn I actually found a youtuber who know what they are talking about and they game !!! Sorry at Frame Chasers but I had to smash the subscribed button and looking forward to future videos
It's embarrassing. Before I knew anything about pcs I thought Gamers Nexus and JayTwosCents were the smart tech tubers.
Jays definitely where you go for water cooling. Anything else, eh.
Me too lol
I wanted to mention this in the comments of Jays video when it came out but got distracted and now it's 2 weeks later and no longer relevant, but the performance uplift from faster memory is platform dependent. Yes, there isn't much benefit on Intel or Ryzen 3D CPUs but on Ryzen non-3D chips, you should match your infinity fabric speeds and run the best latency you can afford because this can be a difference of up to 15% in games. But this ignores the fact that most people still aren't stressing their CPUs in a meaningful way because the GPU or the display are almost always the bottleneck.
Optimum clearly loves and plays games a ton. The guy seems to go hard as f in any hobby he begins but it's obvious he's cracked at shooters.
25%....obviously within margin of error.
LMFAO... When I saw his video my first reaction thought "Buildzoid would go CRAZY if he watches this shit" LOL
Funny thing is can't find any of the comments saying the same thing, might be deleting them 'cause he's very emotional.
yeah hes using a 4090 where as at least jay didn't
look at 7:47 right under where it says Memory Type is your clock speed and it says 2400 MHz not 4800 MHz which it should for your test to be correct and then CL numbers should be 40-40-40-72
Mister, computer game players are not just those who call themselves competitive players. I've been playing for around 30 years and for me what matters is the highest possible resolution and the best image quality that can be obtained at a minimally acceptable FPS number, which varies depending on the game in question. I've been playing at 4K since 2016 and the graphics card is always the bottleneck. That's why I only spend the minimum amount on the CPU and its memories so that these two components don't become a major bottleneck at 4K.The money I save with this option all goes towards the graphics card! And Jay, around 15 minutes in, talks about you!
Bro I watched his video and couldn't believe it. Mind blown smdh
Yup and the video about “Does cleaning your old GPU improve performance” was even worse.
They observed high temperatures and after taking it apart it was then disclosed that one of his friends attempted to service it once before and screwed everything up. Missing thermal pads, loose screws, who knows what else.
Video should have been about a sealed never opened GPU from say 2017 to see exactly what things look like inside, not some mess that someone else already did once before.
Jay does have health problems so maybe that’s why such low effort content but the RAM video blew my mind how bad it was.
you're using a 4090? of course you'll have better performance.
the Gain you get from ram high ram speeds for a someone who is NEW to PC gaming is negligible. 10-15 fps for $100 more is not worth the price especially if youre new to pc why not spend the money elsewhere; that's the point of jays video.
I respect your overclocking videos, you don't have to click bait for views
Best tech yt right here yall. No bs, informative videos that are actually gonna boost your pc’s performance
He is no different from LinusTechTips, all they care about is money & views.
I'll go one step further Jaytwocents channel is not even doing good anymore he has 4million subs and is getting 100-200k views on a video that is terrible not even a quarter of the people clicking on his videos.
I generally can't really handle your videos because of the way you rant. But for once, I would say that this video is quite good with minimal ranting and more to the point which I like. Please make more videos like this.
Break my heart and give me truth!! The real unbiased truth. Just give it to me straight! Thank you Jufes!
I think the biggest thing that is getting missed here is that there is a spot in the middle that probably makes more sense. 6000MHZ RAM with good timings can be purchased for a decent amount less than 7200+. The money saved could be put to upgrading your CPU/GPU, or some other upgrade that you would feel more than having the fastest RAM out there. For a comparison I looked at some fairly standard G.SKILL 32GB/64GB Kits in both speeds and there is about a $40/$70 difference. In budget builds the extra money can go a long way... in systems like yours/mine with everything else maxed out... it might show more of a numbers difference, but can we really see the difference of anything above 280 FPS? That being said i have a a $300+ Corsair Platinum RAM kit... so not sure I should be talking.
I saw Jayz thumbnail and title, chuckled to myself and thought I can't wait until Jufes sees this one lol
Thanks to you Jufes I knew not to even bother clicking the video. 👍
But at 1440p&4k it doesn't really make a difference... I think that was the point of the video he said that at the end.
It does
big L for jay, when he didn't even show the lows or bother with the timings. such a bad vid that even his own commenters call BS
Mate, it’s ridiculous how these guys get so big with the information they throw around.
Good to see you calling it out dude! Keep it up pal! Big support from the UK.
It's even worse with stock market influencers. Even legit guys worth $300 million begged their audience not to buy Nvidia stock at $400/share and today it's $800/share. I know it's unrelated, but if you can't think for yourself in 2024 you're gonna get wrecked in all your pursuits.
💯🎯
@@ryoannazuk949 no one mentioned any meme stocks
@@ryoannazuk949 ohhhhh ok you’re just a super fun troll. what a silly cutie. all you wee little trolls know Nvidia is the third largest company in the world 😘
@@ryoannazuk949 you’re so cute
@@ryoannazuk949 your diction and syntax and reiteration of common Reddit platitudes convey your ignorance, how many shares do you own? What was forward p/e when you purchased? Have you purchased put protection to hedge your investment? What impressed you most about the the last earnings call? The only meme here is your adorable trolly sentiment. Best of luck!
farcry 6 - tomb raider - bf 2042 - warzone and the list goes on from jedec 4800 to 6800 - 7200 with a properly tuned tier 1/2/3 timming , it's a 15 to 35% more perf , plus you get better latency overall from 70- 80ns to 49-55ns , that's why i loved my 12900k delided to 5.30 /4.20e /4.4 ring , i'm rocking the 220 300 fps on both warzone and bf2042 plus a snappy windows experience .... . fun fact an oc12900k 5.40 with a properly tuned rams 6800 -7200 beats the 13900k + 6000 mhz rams stock in tomb raider :D
in my opinion, your the one who deserves the views and clicks these sponsored " techtubers " get.
And the worst part i think Jay believes he is right with this memory speed thing, i don't even know how his video was sponsored by a ram manufacturer and that manufacturer hasn't got mad because Jay dished their fast memory.
I commented and even said Jayz tripping I literally gained 20+ frames going from 6800mhz to 7600mhz 👀
Only worth it at 1080p low detail 🤷🏻♂️
Gained 20 fps on what exactly?
In modern warfare my 7800x3d gained over 100 fps from tuning my fclk, bclk, llc on low-medium, overclocked my 7200mhz cl 34 to 7400mhz, - pbo offset of 25 & a few other settings in bios
z4792 I did a GOTG benchmark I went from 244fps to 266fps I have it on my channel
z4792 also this was on 3440x1440p UW with a 14900k @5.8Ghz E-cores 4.6ghz with a 4090 Tuf
Thanks! As much as I do not enjoy bashing on a particular person. I cannot deny the evidence and the testing you literally did live. I appreciate the edits on youtube and all the information. Everyone, no matter how shady they are will provide good information but eventually the foundation cracks. Money/click-bait becomes the drive for many people.
I absolutely love your no bullshit channel this means allot to have someone like you doing content like this that we as consumers don't get screwed over.
I would really like to do a paid session to help me tune my system but maybe at a later stage as I also got screw over financially but will definitely do this at a later stage this year. Thank you for your content buddy you are awesome.
Let's not forget that Jay was the MAIN fear mongerer that brought forth the entire "fear the POSCap" situation with the release of the RTX 3k series even though POSCaps are higher quality..
Wasn't that more Gamers Nexus, he was so scared about it he made a video while he was on holiday doing his BMX stuff.
@@bigpoppa1234
If you're talking about the video that waas released 3 years ago titled "The Real Problem with RTX 3080 Crashes is NVIDIA'S Paranoia of Leaks" wherein at 0:40 he states: "There's actually been more development to this story, so the addon that you're not going to hear me talking about in the biking video witht he capacitors and crashing stuff; is that Nvidia's released a driver now, which according EVGA, their six cap cards are no longer crashing with that driver change by Nvidia when they were before, so, there's more to this than just the caps"; and later at 4:12 he states "basics of the issue: there's one issue we started researching right before we left, we only had an hour researching it, it's not much time, where the partners are pushing frequency too aggressively like over 2Ghz, and cards will crash when they peak. That's one issue, other one, is the capacitors on the back of the GPU socket being problematic, and I'm not going to talk about the electrical side of it. I'll leave that to other people, but I can talk about like, why this happened, and it's a bit of a partner's fault, a bit Nvidia's fault, and of course partners are very fast to throw Nvidia under the bus, Nvidia, I'm sure will throw partners under the bus..."
"As for the part that's nvidia's fault, a couple things. One of them is that Nvidia's so paranoid about leaks that it's hurting itself", etc. I'm afraid you misunderstood the point of their video entirely.
Steve wasn't saying that the poscaps were the issue, he was saying that their overly aggressive use of them as a filter was a PART of the problem paired with the fact that Nvidia was so paranoid of leaks that they didn't give AIB partners enough time hands on with the drivers and manufacturing specs to PROPERLY design the cards, and technically speaking he was right on the money.
With the RTX 3k release Nvidia told AIB partners that "In order to sell a card of this series you MUST design a new heatsink that cost relative amount in R&D and production as the one we've manufactured" then gave them the information they needed to do it SIX MONTHS later than they normally do giving them an entire MONTH - just one - to figure EVERYTHING they needed to out about these cards, and as you can imagine, that's simply NOT enough time. Adding to this now farced inflation thanks to Ngreedia we had companies SCRAMBLING to try and push out an AIB model in order to try and even stay RELEVANT; and the result was a lot of cards that simply weren't designed properly in a number of ways with one of them being the poscaps.
Don't get me wrong, I'm not defending Steve, or his stance on the issue, but if I'm being frank he has/had a MUCH more quantitative understanding of the issue than most any other "techtuber" that was covering the situation.
Lest we forget, Stephen Burke is one of the only ones of them that actually HAS a background in electrical engineering. The rest of them are pretty much just "self-taught plebs" and I'm in NO way bashing "self-taught" because I'm one of them as well, AND an electrical engineer; but I will say that a proper HISTORY of working with electrical applications and theory helps a hell of a lot more in understanding how this issue came to be than a microsoft certification for software obtained during your tenure as IT.
Forgive any typos, I tried not to spend too long typing this up.
I’ve been eagerly waiting for this video from you. When I watched the first 5 minutes of Jay’s video, I knew that you would have a future roast incoming 😂
My favorite thing is when people are whining about FPS while the CPU and GPU are under 80% usage, and when you say something about memory bandwidth, latency, XMP/tuning, etc, you get the internet equivalent of blank stares and drool.
I've watched that Jay video yesterday... Couldn't help myself and think "oh oh...maybe Jufes has something to say about this"
ram speed matters a lot especially for a badly optimised games. Star Citizen - i714700 ddr5 5000hz - 110fps, i713700 ddr5 8000hz - 200 fps
Jay wanted me to let you know you ran the 4800mhz benchmark at default power limits on the 4090. #Stopthemisinformation
I need help please…
I want to build my first gaming PC.
I play Skyrim mainly with a shit ton
of mods.
I want to be able to play in 4K and use
the new AI mods that are coming out…
I also want to be able to play other games
I’m looking at.
I will need this to also be used as a home/home
business computer.
I’ve seen many say to wait as new and much better
components are coming out soon.
Should I build now or wait?
I want to build an all White system.
Any help would be very appreciated.
Thanks and have a great day.
I'll listen to Jay over you I don't even know you. And he doesn't roast other content creators like you
Ram testing on average framerates was really something else in the realm of hardware youtuber incompetence, holy shit...
any good guides out there you recommend for RAM overclocking?
I’m working on it
@@FrameChasers Oh that's neat.
@FrameChasers wouldn’t playing at 1080p have more impact on the cpu vs playing at 4k? Would that not tax the cpu ram a bit differently than playing at 4k and having the vram on the gpu used more? I seriously don’t know and would be interested in watching or hearing more. Just trying to learn
this video only show 1 game.. while Jay show multiple benchmark..
I'm sorry, but my TEAMGROUPs on 4800Mhz do not cause such a problem as yours. And I have the resolution set to 4K. Personally, for example, I see almost no difference in 4K between 6000Mhz and 7200Mhz. Maybe a lower FPS drop in some places. I tested BF2042, Forta Motorsport, Forza Horizon 5, Cyberpunk 2077. The only thing I noticed is a lot of stability work at 7600Mhz! Anyone who has a board with 4 DIMMs and RAM 7200 Mhz+ has rather invested in the future. And yes I have an i9 13900k and an RTX 4090.
How much difference does it make in 1440p ultrawide (high/ultra)? I know most of these benchmarks are done at 1080p Low/Med to purposely induce a bottleneck but I haven't used 1080p in a decade and I never use Low settings ever. I just care about my real world results.
I run my kit at 6000 CL30. In my limited testing it was like 5% gain for 7200Mhz but a lot more voltage.
@@stevesonger1Did you manually tune your timings and voltages, that’s what Jufes is showing here. Speed+Timings+voltages Tuned= potentially ~25% gains.
@@KiNg0fShReD Yes. At my resolution 3440x1440 I was seeing small increase. I am not disagreeing with Jufes. It is worth it at lower resolutions.
Yea he definitely doesn't play games. Anyone who says how unlikely it is to get CPU bound at a certain resolution doesn't play games. But that is still nothing compared to getting a result where 4800 is faster than 7200 and just running with it.
I've been calling him out for the last 2 years. He's lazy in everything he does. I can't believe more people don't call him out
Would you get the same benefits in 4k and 1440p increasing the ram speed? Are there some diminishing returns in higher resolutions?
@JesusisLordSEEK What cpu/gpu are you running? I have a 7800x3d with cl30 6000mhz ram and was debating on messing with the ram a bit but idk if it's really worth it.
Glad you caught this, thank you
Man I knew you were going to say something about that video. That's freaking awesome.
Hold on. Didn't Jay's chart shows that faster ram gives more fps at 1080p? Your test was also at 1080p. So what is the problem? not shelling for anyone but am I missing the point here?
I've seen 2 or 3 Jays videos, after last one I said to myself no more of that guy... ppl can't be this stupid right? now I just found this, and hope in humanity is slowly recovering. Btw, Im a total noob in PCs, and total pro in humans. Thx for this one! cheers
JayZTwoCents for entertainment, Gamer's Nexus for trusted news and Hardware Unboxed for unbiased reviews, that's what I follow.
I feel like jay started a while back making his employees make the videos, and he just comes in to do a 10 min talk about the graphs they produced while not giving a shit if its true or any good data. He simply doesnt care about the videos anymore and it shows straight through. He made some cool content back in the day but its a long time since I was ablo to watch any of the recent videos released. I dont recall him even mentioning cas latency, only speed. Thats just insane from a ram video
Hi, what’s your opinion about apex encore vs formula? i need a full white build :(
So the video was as bad as I thought it would be, worth waiting for this one
yeah 25% is huge in this game. its multi generational performance numbers looking at it from an out of the box perspective.
this guy and optimum are the only gaming tech youtubers you need
IDK what's worse, to watch Jay on how to choose PC components or just buy a prebuild PC
I'm the only overclocker I know of that actually plays games. I've never got the best results from XMP. It's usually better than stock motherboard settings that go low speed but I usually speed up the cache which you need to do to really take advantage of the faster ram speeds. Tight timings are normally unreliable for gaming. I usually enter the XMP timings manually on the primary's as the board will usually set better secondary timings than XMP. You need balance for the optimum system performance and reliability. Buildzoid is awesome for getting top RAM performance but his reliability sucks for gaming. Go too fast on the memory and the controller will get hot limiting your overclocking capability on the cache and cores. Even if a CPU doesn't throttle excess heat can still hurt performance. It's the price of fast ram that is usually the limiting factor, but generally faster ram will net gains and usually raise the 1% lows for smoothness.
Amd CPU work better better with a increase ram frequency so you will notice bigger performance differences on that platform because amd cpus utilises the speed increases better. So you are more likely to notice a performance difference on a amd platform going from 7000 to 4800. Intel vs Intel platform you might get 1 to 25% differences between each platform. So you won't notice much performance differences.
Yeah, it's pretty bad, but to be fair, there's a _violent_ difference in his audience and your audience. The average normie that watches Jay or Linus isn't tuning their 4090 and direct die cooled 14900ks to get 500 FPS in CoD at 1080p. They're buying a 4060 and a 12600 on a cheap b660 to play Hogwarts and Palworld at 1440p ultra. For that customer they *_won't_* see a real difference in DDR4 and DDR5, let alone 4800 and 7200 DDR5 and it would be stupid to spend twice as much on your memory when it's far better spent towards the GPU.
I build a ton of systems and 99% of them are on a hard budget below $1500 or lower in these times thanks to Bidenomics. For the overwhelming majority of users that can't afford even a 4070ti I have a huge difference in what I recommend and what I buy for myself.
It's easy to go all PCMR and say "don't be poor", but I will do anything in my power to get another gamer off of Gaystation and into PC gaming.
Eh even ltt showed that faster ram among ddr5 gets you appreciable fps difference. That's the video where I settled on a cl30 6000 for the best performance to cost for me. The real flaw with techtubers for normal people is they err on the side of caution since their audience doesn't have 3k to spend on the best of the best components. I even take that into consideration when I use their information.
I thought of you the moment I saw this video, Jufes and couldn't wait for your response.
I don't get it. If I play low quality 1080p on 240Hz+ in CS, yeah the bottleneck is there where the ram makes the diff, but if I play a resolution and game that actually is GPU bound, that's where it's a waste of money... Now I'd say both Jay should be more clear as well as your video... Some disclaimer...
He's right. Why spend more money on faster memory if its not needed? that fps boost from 279 to 350 makes no difference to you whatsoever. Maximize the memory you have, if you need to before bying a faster one. But if you are already in 300 FPS mark, to most gamers it wont make a lick of difference.
As soon as I saw the title of Jay's video I knew shit was going down xD Glad I followed you that other time you called out Boxed on the memory rank debacle.
the benchmark that shows no difference, lets use this one guys.
So I've been mislead? At 4k gaming do I gain alot from anything over 6000 mhz?
The moment I saw Jay's video, I thought that this one is comming.
Same, lol. I was instantly cringing on his behalf, knowing what doodoo he'd just stepped in. 😂
Same lmao
Dufus I was laughing so hard when you lowered the ram to 4800 and you get the same windows experience as I do EVERY DAY ARRRRR with the 7800X3D with memory at 6400. So wished I found your channel before upgrading from Zen 3 to Zen 4.
I want to upgrade to DDR5 now but I also need a motherboard. I was thinking of getting: CMP32GX5M2X7200C34 7200MHz CL34.
What is one of the most high speed RAM stable motherboards out there?
I have a DDR4 system now.
Other specs of my system:
13900K, Arctic Liquid Freezer II 420, 4090, MSI - MEG Ai1300P PCIE 5.0 PSU.
I want to upgrade because I'm still on ASUS TUF GAMING Z690-PLUS D4 motherboard and DDR4 3600 14-15-15-35 TF10D432G3600HC14CDC01. Looks like it's worth it upgrading to DDR5 but Intel might release a new socket soon.
if you're planning on buying a new motherboard to upgrade just wait for the next socket update and go from there then sell your current mobo CPU and Ram as a package
New Jay video! 4090 same performance as a 3090. Plays on low 480p with a Pentium 4.
Thats why I unsubscribed from all of them and started supporting you.
Jay and linus no longer techtubers theyre more like journalists/entertainers nowadays. I've got to say, you hurt my feelings all the time, cuz i have an all amd rig but I Love your content man, its pretty based
I thought Ryzen was the way to go, if I want to build the most high end system, what would you recommend? 14th gen intel? Thanks
Ok, as someone who has been building his own PCs since 2010, i feel really worried, because i don't know what was wrong in Jay's video. I watched it like yesterday and now i got this video in my recommended. I'm not a tech expert or anything. When i choose my PC parts, i just go for the basics. The more the better. Be it MHz, GB, in anything. Seems like i got a lot to learn still...
His numbers didn't add up. He's espousing that more speed on DDR5 at the ones he tested yield no gains, but there's an obvious issue, as anyone who has validated the numbers does not see these results.
Tightening timings and speed bumps typically show clear scaling. In Jay's it did not which means something was wrong with the set up. DDR5 is notoriously tough to tweak, and that's pretty much all that's illustrated in Jayz video.
Its amazing to me that these folks wont consider GPU-bound processes, that use VRAM, will see the least improvements.
I mean, increasing system RAM has no effect on the VRAM - i would have hoped these youtubers would be more honest about this
Havent seen a single video of him, or any other mainstream tuber for months, maybe even over a year now, most of them sold their souls.
I haven't watched Jay since he told everyone to hurry up an buy nVidia cards on black fri
dude there was so many normies that bought that day to lol...... God dam shame......
At least he didn’t tell people to buy amd LOL
@@gildfps true..... I'm pretty sure that would of been against the Geneva Conventions.... Lol......
GN tries to target a more mainstream audience. Your audience doesn't really translate through to any mainstream. Maximum fps really isn't mainstream, since it's so expensive. Spending 4k$ on a pc every 2 years isn't in 99.9% of peoples budgets.
How much difference would I get in performance if I upgrade to 4600 from 3600 ddr4?