If only the series s only problem was teraflops.Series S redeemed?
it does matter but it doesn'tThe "tera"flop does matter, but it's not the whole story. Just because you have the most horsepower in your car, will not mean you can win all the races.
They were pushing it so hard and when it didn't materialize they blamed the "tools" and dev kitsThey are talking like they are not largely participated in creating this teraflops narrative
![]()
That's pretty much the same thing as Carmack pointed out years ago. If you have a choice between doubling the clock speed of a cpu or doubling the core count go with double the clock speed. It makes everything twice as fast where as doubling the core count makes some things twice as fast.it does matter but it doesn't
you see cerny showed 2 GPUs with the exact same TF, he said he prefers one with lower CUs and higher clocks because it will be faster, while the bigger and slower GPU is hard to use all his slow CUs
If only the series s only problem was teraflops.
They were pushing it so hard and when it didn't materialize they blamed the "tools" and dev kits
![]()
If only the series s only problem was teraflops.
I mean, this tweet is just as wrong, if not more wrong, than the teraflop argument ever was.
There's nothing about the PS5 that makes it "one of the most revolutionary home consoles ever designed".
Not until that Ratchet and Clank PC thread anyway.And I don't think anyone lauded it as such so he would have lost that bet.
Not until that Ratchet and Clank PC thread anyway.
Have mentioned this for quite awhile but they really wont want to talk about TFs when the PS5 Pro launches as its going to look like a letdownTeraflops are relevant when comparing processors (of the same architecture), but not overall systems. I think that is the lesson that should be learned going forward here. PS5 and XSX have different internal designs. Also different APIs. But yes, using teraflops as a marketing tool for console is completely useless.
Teraflops are relevant when comparing processors (of the same architecture), but not overall systems. I think that is the lesson that should be learned going forward here. PS5 and XSX have different internal designs. Also different APIs. But yes, using teraflops as a marketing tool for console is completely useless.
Have mentioned this for quite awhile but they really wont want to talk about TFs when the PS5 Pro launches as its going to look like a letdown
If teraflops don't matter what metric should we use?
I was joking. But sometimes it's hard to tell if people are serious when they say the game was better on PS5 because of its mystical I/O power.lol.....I must have missed those posts.
![]()
It's all about the magical SSD.
The point is you shouldn’t use any single metric like TFLOPs or GHz to evaluate a system.If teraflops don't matter what metric should we use?
But it’s been three years and i have not seen anything revolutionary inspired. It’s been the same old stuff from the ps4 era.They were pushing it so hard and when it didn't materialize they blamed the "tools" and dev kits
![]()
Yes. Teraflops isn't really a performance metric. It's a theoretical measure of throughout based on the type of floating point math you're doing. More likely than not the same processor will have different levels of throughput depending on whether you're doing single-precision or double-precision arithmetic and actual performance is impacted by power draw, heat, frequency...all of those things that affect efficiency. You have to look at it in context.That's good. I've never depended on TF to buy a GPU. I look at benchmarks comparing frame rates at specific resolutions. That's how I'll judge the Pro as well.
Actual gaming benchmarks.
I know about Cerny's explanation (even after 3 years). The reason is because in the previous generation the the PS4 had 18 CUs and PS4 Pro 36 CUs. The game engines were built and updated to work ideally with 18 CUs - and 36 for higher visual settings. This why, to my understanding, in the PS5 the 36 CUs at higher frequency was wiser, because anyway the existing game engines wouldn't be able to take full advantage of the even more CUs. Only later in the generation with new or updated game engines. But with higher frequency it's much easier to do it from the beginning in the console lifetime.it does matter but it doesn't
you see cerny showed 2 GPUs with the exact same TF, he said he prefers one with lower CUs and higher clocks because it will be faster, while the bigger and slower GPU is hard to use all his slow CUs
And teraflop is only a constituent of a GPU's horsepower to begin with. There are other throughputs but DF seem to intentionally avoid mentioning the other pieces of the power puzzle since the beginning of the generation.The "tera"flop does matter, but it's not the whole story. Just because you have the most horsepower in your car, will not mean you can win all the races.
They aren’t really upping the tflops though. Increased tflops is just a byproduct of a more powerful system through wattage increase/more cores/higher efficiency/superior techSigh. Tflops do matter. If they didn’t, Sony would be releasing a 10 tflops ps5 pro with more secret sauce io and 50 gbps ssds instead. No, they will still up the tflops because that is literally the computing power of the gpu.
What ms did is create a gpu with obvious drawbacks and paired it up with a bizarre split ram solution. It’s a poorly designed console that needs to be criticized and studied to ensure same mistakes aren’t repeated. However, one poorly architected console shouldn’t be used to say that tflops don’t matter.
It’s not that Sony have secret sauce, they just aren’t bogging down their system with layers upon layers of sub-optimal abstraction. Sony’s consoles allow easier access “to the metal” while on Xbox the whole OS is virtualised and encrypted through their highly inefficient (relatively speaking) hypervisor.I think it does matter, and not to be a contrarian but I do think it could be the API. Something doesn't add up about the Series X. The CPU is even faster. The ram is faster. Could be due to the ram being split but I am not sure. The PS5 isn't doing something exotic either, other than having a higher GPU clock but I have never seen this replicated with PC cards. If Sony really does have some magic custom hardware then they would say.
We are not experiencing that because we have been plagued by cross gen games for 2 fucking years and main Sony studios are yet to show its true Power.They were pushing it so hard and when it didn't materialize they blamed the "tools" and dev kits
![]()
22-23tf is a letdown in your opinion?Have mentioned this for quite awhile but they really wont want to talk about TFs when the PS5 Pro launches as its going to look like a letdown
Well said. Xbox wanted to have a bigger TF number, regardless of what needed to be sacrificed in order to achieve it.Sigh. Tflops do matter. If they didn’t, Sony would be releasing a 10 tflops ps5 pro with more secret sauce io and 50 gbps ssds instead. No, they will still up the tflops because that is literally the computing power of the gpu.
What ms did is create a gpu with obvious drawbacks and paired it up with a bizarre split ram solution. It’s a poorly designed console that needs to be criticized and studied to ensure same mistakes aren’t repeated. However, one poorly architected console shouldn’t be used to say that tflops don’t matter.
The Series S GPU being that small means it should've been close to the PS5 in terms of frequency and a 192 bit bus so would be much closer to the XboneX with regards to fill rate and bandwidth.Weak GPU and they fucked up with the memory pool size and speed.
A completely unforced error, a silly own goal.
Should have been at least a 6TF GPU + VRAM at least at par with the One X.