• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

SiSoftware Tests the Ryzen 5 7600X, Ryzen 7 7700X and Ryzen 9 7950X

Speaking of memory bandwidth... Would a large 256-512 MB L3 or L2 cache next to the CPU cores help or the bottleneck in the execution is in the cores themselves?
I see that the 96 MB cache on the Ryzen 7 5800X3D doesn't help for anything outside games?
Not much, no - except for in server/HPC workloads. Much like AVX indeed!

"The first reliable benchmark figures of AMD's Ryzen 7000-series CPUs have arrived

"Also note that the graphs for the Ryzen 5 7600X have typos"

So reliable they can't get the four-digit model number of just-announced products correct.
Hey, not all reliable sources have copy editors :P This does seem like one of those somewhat rare cases where a reputable source has somewhat lacklustre quality control in terms of spell checking though.
 
Comparing the 7000 to 12000 right now is bad. RL is right around the corner.
 
Comparing the 7000 to 12000 right now is bad. RL is right around the corner.
... and they are somehow supposed to compare them to unannounced, entirely unofficial products that they most likely don't have?

People keep saying this, as if it makes any kind of sense as an argument. You compare what exists with what exists - today. Then, when something new comes into existence, you compare them again. You don't wait for the next thing - then you'd never be able to compare anything at all. And, crucially, having access to early silicon for one series of CPUs - announced, with an impending launch date - does not necessarily mean you have access to early silicon to another - unannounced, more distant - series.

If you have a meal and talk about its quality, do you compare it to things you have already eaten, or do you talk about how it might compare to what you'll be eating tomorrow?
 
... and they are somehow supposed to compare them to unannounced, entirely unofficial products that they most likely don't have?

Both are unreleased but they do exist. If they cared, they could find a way to find a 13th gen CPU and compare apples-to-apples!
 
Both are unreleased but they do exist. If they cared, they could find a way to find a 13th gen CPU and compare apples-to-apples!
You're joking, right?
 
... and they are somehow supposed to compare them to unannounced, entirely unofficial products that they most likely don't have?

People keep saying this, as if it makes any kind of sense as an argument. You compare what exists with what exists - today. Then, when something new comes into existence, you compare them again. You don't wait for the next thing - then you'd never be able to compare anything at all. And, crucially, having access to early silicon for one series of CPUs - announced, with an impending launch date - does not necessarily mean you have access to early silicon to another - unannounced, more distant - series.

If you have a meal and talk about its quality, do you compare it to things you have already eaten, or do you talk about how it might compare to what you'll be eating tomorrow?
It is a waste of time, that what it is.

If you comper burgers and visited 2 brands and on your way to the 3rd, please wait till you finish the 3rd Burger before expressing your conclusions.
 
Isn't Ryzen 5 supposed to be 7600X? why in charts its called 7760X??
 
Yet there are some major uplifts in performance when using it, by limiting it's availability like intel has you're going to see a limitation in developers wanting to even play with it.

Due to the fact that the RPCS emulator sees a major uplift in using it tells me there's a lot of hidden potential in it.

Also, intel wants to lock it into their server market only because of $$.

I've had avx 512 since the skylake 7820x... there's a ps3 emulator that uses it. Games don't use it since they don't need the SIMD math performance -- they have a GPU attached. Some video encoders and scientific simulators use it, but again -- you're competing with GPUs -- really anythingsavx512 is really great for is much, much faster done on the GPU.

Zen4's secret weapon isn't AVX512... it's 3d cache, and amd is waiting to release that only because of $$.
 
Last edited:
Isn't Ryzen 5 supposed to be 7600X? why in charts its called 7760X??
As I pointed out, it's a typo. They messed up the Intel chip in those same graphs too.
 
It is a waste of time, that what it is.

If you comper burgers and visited 2 brands and on your way to the 3rd, please wait till you finish the 3rd Burger before expressing your conclusions.
What? No. That's a terrible approach. You compare the first two first, to hash out your impressions and thoughts, identify points of comparison, qualities of each, etc. Then you bring that knowledge to the third, and let that inform the final comparison (including any entirely new findings). Waiting until you've had all three - unless they're right after each other with no gaps - muddies your recollection of the two first ones, making the final comparison more difficult and less accurate.

And, once again, by that logic you would literally never be able to compare anything, as there's always something new coming. Always.

@ARF: the reason I hope you're joking: Not only are you portraying a fantasy-land idea of how people get access to pre-release hardware ("If they cared, they could find a way" - tell me, do you have a 13th gen CPU? Could you get one if you cared enough?), but you must understand that prematurely leaking something like this, before the product is even announced (unlike Zen4, which is launching in three days) would seriously jeopardize the very same relationship they might have with Intel that might grant them access to such hardware for development purposes.
 
Last edited:
Well, I think in both cases the NDA is violated.
That depends on the terms of said NDA, if there is one at all, or if there is some other kind of agreement. But even without that, what matters isn't so much the potential contractual violation as the violation of trust. If you trust someone and they screw you over, you don't trust them again.
 
What? No. That's a terrible approach. You compare the first two first, to hash out your impressions and thoughts, identify points of comparison, qualities of each, etc. Then you bring that knowledge to the third, and let that inform the final comparison (including any entirely new findings). Waiting until you've had all three - unless they're right after each other with no gaps - muddies your recollection of the two first ones, making the final comparison more difficult and less accurate.

And, once again, by that logic you would literally never be able to compare anything, as there's always something new coming. Always.

@ARF: the reason I hope you're joking: Not only are you portraying a fantasy-land idea of how people get access to pre-release hardware ("If they cared, they could find a way" - tell me, do you have a 13th gen CPU? Could you get one if you cared enough?), but you must understand that prematurely leaking something like this, before the product is even announced (unlike Zen4, which is launching in three days) would seriously jeopardize the very same relationship they might have with Intel that might grant them access to such hardware for development purposes.
The only reason and I mean only reason why some want to perpetually compare something that is unreleased is when they are routing for that unreleased thing in hopes that it will narrow or out right beat what they are not routing for. Some here want Intel to always be ahead so they cast constant whataboutisms. Same goes for those who like AMD. Its a nuisance but ever present ‘feature’ of internet comments.

Hey they didn’t compare X to Y so I can go to sleep happy in my thoughts tonight.
 
That’s the obvious and best choice, after that we can choose between Raptor Lake, Ryzen 7000, Alder Lake and even Ryzen 5000. Couldn’t care less about platform longevity, I always try to choose my CPU wisely for my needs.
I don't agree. Platform longevity is a major factor. AM5 is at the start of the journey. Raptor Lake is the final hurrah for the platform. While Raptor Lake has a more mature platform with conceivably fewer teething issues, if one is putting together a new build then AM5 offers the 7000 generation and several further generations on the same platform which from an economic and eventual stability perspective is the better choice. While I always wait for reviews, those advocating for people to wait for Raptor Lake before jumping have discounted the wider view.
 
I don't agree. Platform longevity is a major factor. AM5 is at the start of the journey. Raptor Lake is the final hurrah for the platform.

Say, doesn't Intel change platforms every other year?
 
That depends on the terms of said NDA, if there is one at all, or if there is some other kind of agreement. But even without that, what matters isn't so much the potential contractual violation as the violation of trust. If you trust someone and they screw you over, you don't trust them again.

Not to mention that some of the Zen4 cpus where being sold before the official launch date, it would be hard to enforce an NDA on something that they can buy themselves.

 
Probably AMD use same CCDs for servers and users
That sounds likely. What confuses and even baffles me is why Intel included it in the Golden Cove cores - when each CPU is monolithic and is not reused among multiple segments. If the hybrid architecture was to maximize core counts in a given die area, why in thunder did they add the space-hog of AVX-512?!?

However, using identical chiplets across all market segments could boost adoption of new SIMD instruction sets if/when Intel starts doing the same thing.

I kinda hope Zen 4 comes in smaller chiplets as well, for the low end. However, monolithic dice (recycled from mobile) seem more likely there, to keep package costs down.
 
What? No. That's a terrible approach. You compare the first two first, to hash out your impressions and thoughts, identify points of comparison, qualities of each, etc. Then you bring that knowledge to the third, and let that inform the final comparison (including any entirely new findings). Waiting until you've had all three - unless they're right after each other with no gaps - muddies your recollection of the two first ones, making the final comparison more difficult and less accurate.

And, once again, by that logic you would literally never be able to compare anything, as there's always something new coming. Always.

@ARF: the reason I hope you're joking: Not only are you portraying a fantasy-land idea of how people get access to pre-release hardware ("If they cared, they could find a way" - tell me, do you have a 13th gen CPU? Could you get one if you cared enough?), but you must understand that prematurely leaking something like this, before the product is even announced (unlike Zen4, which is launching in three days) would seriously jeopardize the very same relationship they might have with Intel that might grant them access to such hardware for development purposes.
The RL lunch is so here, that's why you better wait. If it wasn't so imminent I would totally agree with you.
 
I heard the Raphael chips don't lose much when you turn down the power and are very efficient. As powerful as these chips are I'd probably never need full power anyways right now. I usually crank the juice up and o.c. when it's getting close to upgrade time.
 
AMD has also improved the inter-thread/core latency in the same module, by a not insignificant amount.
I think this is the best under-reported item.
 
Those aren't reviews.
Look what they have posted and what they called it.
Those aren't reviews.
So, you didn't visit their website. Let me help you and your welcome. :toast:
SiSoftware screenshot.jpg
 
Back
Top