AMD Ryzen 9 9950X3D2 Review: Twice the 3D V-Cache

[fastbyjoe]

Thanks for the Adobe products tests!

 

[themastergoose]

Is the 9950x3d2 in a market all by itself, or are other products sold that compete with it?

And did you miss they said "productivity VALUE"? Not just "productivity". That "value" is crucial context. The 9950x3d2 is triple the price but doesnt offer anywhere near triple the performance in productivity for that price.

Halo products have never made sense from a cost vs performance standpoint, but that was never my point.

My issue is the entire article conclusion says for productivity the Intel CPU is unbeatable. I’d argue it isn’t, because most people aren’t going to buy one system for productivity and another for gaming.

The value comes from having a single system that actually doesn’t really have any performance compromise in EITHER workload rather than needing two distinct systems.

 

[Vanderlinde]

The extra power comes from keeping the SRAM in an active state. It's a over the line of a CPU; that likely offers something in certain workloads, and is cheaper then a Threadripper.

Still a bad-*** CPU.

 

[Squid Surprise]

Halo products have never made sense from a cost vs performance standpoint, but that was never my point.

My issue is the entire article conclusion says for productivity the Intel CPU is unbeatable. I’d argue it isn’t, because most people aren’t going to buy one system for productivity and another for gaming.

The value comes from having a single system that actually doesn’t really have any performance compromise in EITHER workload rather than needing two distinct systems.

The GPU tends to be the limiting factor in most games… if you really had unlimited cash for just 1 system, you’d buy a Threadripper 9980 with a 5090… or the Threadripper Pro 9995 if you’re going all out I suppose…

 

[Jyrkz]

Somebody didnt read the whole article. I circled it in red for you.
Gaming benches..

Somebody didnt read the whole article. I circled it in red for you.
View attachment 90847

Gaming benches ofc..

 

[AdamNovagen]

Sounds to me like AMD might be trying to play the superior product so superior pricing game. Maybe people will buy this CPU, but maybe not. If Intel is significantly cheaper on a price/performance ratio, then AMD only stands to loose to their own corporate hubris.

Corporate hubris, IMO, bit Intel, and many other companies, in the butt; however, AMD, or so it seems, has failed to learn the lesson.

We saw this kind of back and forth between AMD and Intel back in the days of Intel's Core 2. For me, its a big turn off even though I'm primarily an AMD buyer. ATM, I'm not compelled to buy this latest iteration of the part, and I hope lack of interest in the consumer realm causes AMD to finally see that name does not outweigh price/performance ratio.

This summarizes the whole situation imo. Been an AMD fan since 3rd-gen Ryzen, but these last couple years have mostly just been a giant middle finger from AMD. I have nowhere left to go at this point.

 

[HomeyBroDude]

I WAS looking to replace my 5950x. I think I can wait a while longer.

 

[Biostud]

Really really poor form from HUB testing like it's a normal consumer cpu. This is a poor man's HEDT replacement for TR.

If you want sensible testing look at Phoronix and and see what software actually thrives on cache. For example OpenFOAM fluid sim software. The people that buy these know exactly why they need more cache and could care less one bit about utterly useless tests on Cinebench, compression and gaming.

Uhm,techspot is a gaming site and phoronix is a Linux/work site, hence the difference in benchmark suites.

But I agree, that this is a "poor man's" THREADRIPPER, for people who has these kind of workloads and likes to game.

 

[mbk34]

It would seem far better to just have the single CCD and double the V cache. I'm sure AMD do testing to see what's optimal for the cost but such a product might spark some interest. Then call it a 7800 X3Dx2?

 

[JamesBlond]

Would be interesting to do a 2 year burn in process running RandomX 80% cores, max speed and see how much variation there is in calculation, if it maintains 98% accuracy overall through the first 2 months test its good

 

[mbk34]

I'd be more interested to see how a 9800X3D2 would perform. Having extra cache made a huge difference in gaming when the 5800X3D was introduced so how would doubling the amount of cache again affect things? Is there any way of seeing whether the existing cache gets totally used and the processor resorts to swapping with the RAM?

It would also make an interesting article saying how programs pick which processors to run on - this applies to processors with hyperthreading, or with E & P cores or with dual CCD's where one CCD doesn't have additional cache. What happens when the program runs on a processor with 2 cores or 16? Is it handled by the OS or can the program decide?