- Mar 3, 2017
- 1,747
- 6,598
- 136
If you re-read the corresponding study carefully, you will find that these were in fact 9 out of 10 dentists, not AMD enthusiasts.For everything else Intel 13600/14600 CPUs are generally regarded to be the best universal CPUs with excellent price to performance ratio. 9 out of 10 AMD enthusiasts will confirm that.
Not in chiplet configuration.So, what "is already established" is in fact that the Zen 5 core is consistently better than the Zen 4 core at low power levels in integer workloads (and at high power levels too).
If default TDP of 96c Turin vs. 96c Genoa is raised as stated in this rumor, then higher supported memory clock alone would already be a sufficient reason to do this.
Then it'll be Vieja Lake.So if Nova Lake releases in 2035 I guess it's not delayed.
I guess you just blanked that every 13th and 14th gen CPU with 65W TDP or higher is part of Intel's massive debacle. Today, it doesn't matter what price to performance ratio any of those products offer, no one should be considering buying them.I just checked some reviews and in fact 9700X has pretty similar performance as 14600K, I thought 14600K was a bit better... It is still cheaper and better in multithreaded applications though.
So, what "is already established" is in fact that the Zen 5 core is consistently better than the Zen 4 core at low power levels in integer workloads (and at high power levels too).
You are correct if you are saying that nobody has presented similarly detailed measurements of performance over core power for Granite Ridge vs. Raphael yet.Not in chiplet configuration.
Why are you so funny? Why would AMD enthusiasts confirm 13600/14600 are better value for gaming when it's not even clear if they will work properly in the long run without degrading? Even without degrading, I see no reason for AMD enthusiasts for recommending Intel products. It's not as if AMD doesn't have its own competing products against 13600/14600 SKUs.For everything else Intel 13600/14600 CPUs are generally regarded to be the best universal CPUs with excellent price to performance ratio. 9 out of 10 AMD enthusiasts will confirm that.
Section "Cache Setup" of Chips and Cheese's Ryzen 9950x analysis says otherwise.BTW.
a moment ago I came across this information on another forum
"Reviewers have already caught that communication between the core and the cache in zen 5 has a 40% higher latency on zen 5 than zen 4, "
From a linked in post we know that only 16 people worked on the bobcat / jaguar based PS4 (& helped deliver AMD out of bankruptcy )I wonder how many people worked on the ZEN5 architecture project, how many core designers?
I had not seen Huangs part 2 analysis. Once again it appears that he once again single handedly has delivered the best analysis to date. Wish he would produce a Granite Ridge entry.(David Huang's Strix Point analysis part 2)
You are correct if you are saying that nobody has presented similarly detailed measurements of performance over core power for Granite Ridge vs. Raphael yet.
AMD went from a 500 mil market cap at around that time to over a 100 bil market cap by August 2020. If they still only had a literal handful of designers and engineers working on Zen 5 at that point, shame on them and shame on their CEO, but I very much suspect that is not the case, lol.From a linked in post we know that only 16 people worked on the bobcat / jaguar based PS4 (& helped deliver AMD out of bankruptcy )
It just shows where their priorities are. Their best people are most likely working on server and AI products.If they still only had a literal handful of designers and engineers working on Zen 5 at that point, shame on them and shame on their CEO, but I very much suspect that is not the case, lol.
And in those benchmarks large part of power was consumed by the IODie, eating into core power budget. Also vs Zen4 scores posted by the users here iirc. Since Turin has different IODie it might behave differently. I am not sure if Turin gets new IODie or is reusing Genoa design.I had not seen Huangs part 2 analysis. Once again it appears that he once again single handedly has delivered the best analysis to date. Wish he would produce a Granite Ridge entry.
I was speaking of 9950X vs 7950X and 9700X vs 7700X/7700 low power MT Cinebench scores (some of which was shown here by the ES user). Its not anomalous, its universal, that perf/w , at least in Cinebench r23 ( and presumably all other fp heavy apps, at least) that chiplet based Zen 5 drops to and even below chiplet based Zen 4 levels at 45W and below TDPs. The data is freely available and it is not scandalous or hyperbole to point it out.
If you are interested about architectural details instead of general apps performance recommended sources for Granite Ridge are Chips&Cheese and Alex Yee's Zen5 AVX512 teardown. For Strix Point add David Huang's analysis into the list. Fact check whatever you read on any forum against these.BTW.
a moment ago I came across this information on another forum
No. Just no.
For gaming AMD 7800X3D is the best, it has no competition beside other more expensive and for gaming unnecessary AMD 3D CPUs.
For everything else Intel 13600/14600 CPUs are generally regarded to be the best universal CPUs with excellent price to performance ratio. 9 out of 10 AMD enthusiasts will confirm that.
Even worse. The 13900k and the 14900k should not even be in the picture due to stability , do the 9600x and the 9700x should also be in the picture. And to be fair on cores, the 9900x and the 9950x
I was speaking of 9950X vs 7950X and 9700X vs 7700X/7700 low power MT Cinebench scores (some of which was shown here by the ES user). Its not anomalous, its universal, that perf/w , at least in Cinebench r23 ( and presumably all other fp heavy apps, at least) that chiplet based Zen 5 drops to and even below chiplet based Zen 4 levels at 45W and below TDPs. The data is freely available and it is not scandalous or hyperbole to point it out.
Yes, I don't recall that these Cinebench studies (from the ES user, and from some Chinese videos IIRC) showed actual core power, or that deducing core power from the given info was possible with reasonable certainty.And in those benchmarks large part of power was consumed by the IODie, eating into core power budget. Also vs Zen4 scores posted by the users here iirc.
It is a new (at least: updated) IOD which supportsSince Turin has different IODie it might behave differently. I am not sure if Turin gets new IODie or is reusing Genoa design.
Lisa Su own specialty is semiconductors design, not CPU design wich is surely delegated to Mark Papermaster with Mike Clark just below him.
Ironically Lisa Su s competences would had more suited as head of TSMC.
No. Just no.
For gaming AMD 7800X3D is the best, it has no competition beside other more expensive and for gaming unnecessary AMD 3D CPUs.
For everything else Intel 13600/14600 CPUs are generally regarded to be the best universal CPUs with excellent price to performance ratio. 9 out of 10 AMD enthusiasts will confirm that.
I guess you just blanked that every 13th and 14th gen CPU with 65W TDP or higher is part of Intel's massive debacle. Today, it doesn't matter what price to performance ratio any of those products offer, no one should be considering buying them.
A Kraken 242 SKU without NPU or with reduced NPU throughout would make sense, yes. Specially given it's a more budget offering.Throwing my current guess for Zen 5 mobile naming, will be interesting to see next year what it actually is. Either way, it's going to be a mess.
View attachment 106198
It's much simpler than that. They aim at similar markets with similar goals so there's a floor on what you can do. Same with athletes that differ by less than 0.1s wins, despite from being a different background, language, and culture.how do both companies know in advance the average scores of each other so they are so similar?
on cutting edge CPU archs? the most complex devices on planet earth (and manufacturing lithography ASML etc etc)
some insane espionage going on
Its free extra performance for everyone, just still no reason for gamers and other casual users to upgrade their Zen 4 build. So you get better performance upgrading Windows to 24H2 than buying a new CPU.
So with the new Windows 11 build, you are getting a 10% boost on avg for Zen 4 & 5. Bad news is that it doesn't make Zen 5 any better than Zen 4 in gaming, because their relative position doesn't change.