If I have the 3034 error, do you still recommend removing blu-ray drive and putting pressure on the RSX clamp to check if it's BGA? If I kill the chip even more, it's not the end of the world since it's already giving me 3034, but I'd rather not do it until you confirm this is still something you recommend trying (I read a forum post of yours from 2021 mentioning this, and I know how much has changed since then in terms of YLOD knowledge)
700/800-850 should be enough. After 800-850MHz VRAM, there is diminishing returns, so it's not worth it.Hello!
I recently upgraded from the 2001A Slim to the 2501A Slim for best performance and reliability. I'm interested in putting 4.91 EVILNAT BETA 8 OC firmware as it's publicly available but I'm not sure how to set the MHz I desire and not sure if someone has suggestions.
I looked in this forum a few days ago and someone mentioned that for the CXD5300A1GB RSX that I have in my 2501A unit can do 700/900 good. Not sure if this is a good combination or is there a better combo for my RSX or if I should go with that combo!
If someone knows how to put the file together so that it's ready to go in my console or help out adjusting the appropriate MHz from someone I would really appreciate it!
Nice video !Once again, thank you very much @Mitsu™ for the easy access to the OC firmwares!
I have posted a comparison benchmark YouTube video on Gran Turismo 5 1080p between 500/650 and 700/850 on a launch CECH-2001A Slim with less than 11 days of runtime. This overclock should be a realistic number for most PS3 consoles (other than 90nm RSXs), before encountering diminishing returns.
Looks like I can't post the video just yet.
7 FPS more isn't diminishing returns. Those are gains. None of you have done Overclocking on pc and it shows. Overclocking isn't some magic bullet that gives 50 to 60 plus extra FPS. The gains are huge for PS3 and it's a fact. But that's cause 30 FPS is the target, and games that do 60 FPS is the target. and these OCs reach or nearly reach the target. And every game will react to OC differently. As long as it's mostly gpu bound, the OC will do a lot more than others. I'm sure every game is suffering due to trash cpu. Even the gpu demanding ones. But the gains aren't Diminishing, they just aren't supposed to be a whole new graphics card of uplifts. In my tests, every 50 core translates to 3 FPS or so. Games are dynamic. You can stand still and benchmark 1 area but that's not how the whole game will behave. Sure benchmarking certain spots gives you a great apples to apples, but in other more demanding areas it could dip. And certain scenes or areas could easily and so become cpu bottlenecked. Until we can actually get a cpu overclock, we won't know just how much these gpu overclocks do for the RSX. I'm big into Overclocking on pc. And OCs on pc can gain you anywhere from 6 to 15 FPS. Which is a lot when one is targeting more stable higher frames or targeting 60 FPS.700/800-850 should be enough. After 800-850MHz VRAM, there is diminishing returns, so it's not worth it.
Check here for firmwares :
https://app.mediafire.com/0uxvp5ismnjwm
Nice video !
(Btw, you were right about diminishing returns, sorry ^^')
what kind of plugin or app do you to know which shows the mhz in left side of screen can you share the plugin or app plzI checked it with several games and what the duck test gives is confirmed
Look at GT6 1080p, the difference between 700/900 and 800/1000.
One example among others.
View attachment 43464
Speaking of games that aim for 30 FPS, 700/800-850 is more than enough for them.
Again, if despite that a game run badly, it's not a GPU issue. It's either a CPU-bottlenecked title or it just suffer from $hitty optimization.
So I confirm and sign, buy a CFW-COMPATIBLE 25XX no matter its RSX revision, and enjoy.
700/800-850 is more than enough for most of the games, with an insignificant impact on lifespan. After that, you'll still have gains, but diminishing returns.
So, it's useless to stress the hardware further.
The GPU alone can't save everything !
I'm not the only one saying this btw, @Ketxxx posted about it several times.
https://www.psx-place.com/threads/p...peeds-ps3-cfw-only.36801/page-115#post-392074
I admit that I was a bit doubtful at first, and that's even why I did tests myself to see, and he is right.
For further meaningful improvements, what we need is a CPU OC, I agree.
If you have concrete examples, results, to prove that mine are false, go ahead. I have no trouble admitting that I am wrong when that is the case.
On the contrary, it will help our research.
Well, four of us come to the same conclusion now, @Ketxxx , @RIP-Felix , @SimRacer778 and me.
No that's not the case. I've tested over 60 games. They all have varied results. Y'all don't even have the right tools to test 1% or 0.1% lows. None of you know what you are even talking about. Once again, game to game basis OC can act differently accross all games, some huge some not too huge. All games react differently, due to different engines, resolution target, game settings, and how much the cpu does. And no, I'm only selling one. The rest are mine.
Oh okay, you convinced me.
The hardware doesn't get stressed at those frequencies. Again you don't know what you are saying. Higher frequencies doesn't do a thing to the hardware. The temperature doesn't increase at all due to OC. No extra voltage is going through it, once again just like a pc! Frequencies don't degrade or damage hardware, voltage and temperatures do. And just like I said, overclocks will greatly vary from game to game. I played through all of Transformers fall of cybertron at 1080p and it kept a 30 FPS average through the whole play through, and only dipping down to 24 at extreme scenes with tons of things going on and also increased its graphics and put 16x anisotropic filtering. While it performed worst at lower resolution. These games are so nerfed in res that cpu gets involved too much. If games were to be rendered at an actual 720, 900, or 1080p more work goes to the gpu than the cpu. Which TFOC proved. It ran great while running at a crisp 1080p. Resolution being so ass on the PS3 is a big reason why games perform so poorly, the cpu gets a bigger bottleneck with its low clocks and optimization and architecture.Oh okay, you convinced me.
We're all wrong then, we suck, sorry
You are the only expert here, it seems.
While the exact FPS increase might vary slightly from game to game, the overall trend of diminishing returns is consistent. For example, one game might show a 15% improvement from a specific overclock, while another shows a 12% improvement, but both will likely show much smaller gains beyond 700/850.
The practical impact of these variations is often small. For most users, the difference between running a game at 700/850 versus higher OC is not significant enough to justify the increased risk and potential hardware stress.
The risk free OC myth, huh ?
...
1 - Install WebmanMod.
dev_hdd0/tmp/wm_res/VshFpsCounter.yamlI'll try this.
Unfortunately, no. It's the CPU that does nearly everything IIRC.
Higher frequencies when the silicon can support it does not cause wear. Seriously you don't know what you are talking about at all. You're regurgitating the same dumb things, the uneducated keep spewing. Voltage and high temps is what kills components. Hell and voltage technically doesn't cause the Wear, it's the heat due to higher voltage, or heat in general. And I literally said, 720,900 and 1080p. And FOC at 1080p looks like the damn pc version on the PS3. It is crisp. And I maintain 30 FPS through the whole game. While it would fold hard without the OC. It did 15-18 FPS average when I tested at 650/800. Why? Cause my brother has a ps3 fat and he decided to play through it too. Even at 15 to 18 FPS at all times. He didn't care cause he wanted to replay through it and give me data at the same time. I played through it at 850/950. Then I tried 950/1000 and the game got even faster. And when the target is 30 FPS at higher res. Every single gained frame means a lot. Stop gathering what like 2 or 3 people say here and say it's fact. Cause it isn't. Especially when y'all are testing 2 or 3 games at most. And coming up with such asinine conclusions, and still being full blown wrong. I love your videos though and love your passion for this. I respect you big time, you help a lot of people here. But we gotta keep the misinformation to a minimum.
At least we agree on that
It works as intended, for me.
Ok thanks for the concession.View attachment 43476
At least we agree on that
I don't particularly want to continue debating this subject with you, it's useless.
It works as intended, for me.
Sorry, but this is straight up wrong. The temperature does increase when you increase the frequencies. My slim for sure does run hotter on 900MHz than on 500 Core.
No it doesn't. At all. Stop. I did a 5 hour test with 500/650 and then did 5 hours with 950/1000 in crysis HD. Temps were identical. The amount of cap is crazy. I also did an XMB temp test. And again. Identical. Nothing changed at all.
