AMD Ryzen 7 5800X3D

[nicamarvin]

Linux Benchmarks at Phoronix is Up..

As expected the HPC, 3D Fluid Dynamics and Deep Learning Performance is Off the Charts

 

[GURU7OF9]

In a review of the new "GAMING focussed AMD RYZEN R7 5800X3D " why would you choose to test 38 applications and only 10 games?
This defies common sense and logic !
It should have been 38 games tested and only 10 applications !
Maybe I missed something ? Confused ?

 

[zx128k]

In a review of the new "GAMING focussed AMD RYZEN R7 5800X3D " why would you choose to test 38 applications and only 10 games?
This defies common sense and logic !
It should have been 38 games tested and only 10 applications !
Maybe I missed something ? Confused ?

Linux is likely the OS.

 

[GURU7OF9]

Nope, it says in test setup used win11.
Maybe somebody just wants to see what apps will benefit from 3D V cache ? Or highlight the fact that 12900k will crush it at apps !
I still wonder what the logic is behind it?
It just doesn't make any sense to me other than, think coincidence - think again !

 

[zx128k]

Nope, it says in test setup used win11.
Maybe somebody just wants to see what apps will benefit from 3D V cache ?
I still wonder what the logic is behind it?
It just doesn't make any sense to me other than, think coincidence - think again !

Must be interested in productivity applications and some gaming.

 

[nicamarvin]

Linux is likely the OS.

He is Clearly not talking about Phoronix....

 

[R0H1T]

Or highlight the fact that 12900k will crush it at apps !

What apps?

 

[zx128k]

He is Clearly not talking about Phoronix....

PPL work and play some games on Linux. Or dual boot. Many poeple are not solely focused on gaming. Some people want to know more about how it affects productivity programs. Each review caters to its demographic. If purely interested in gaming review there is lots of sources to find more information.

 

[Aquinus]

Linux Benchmarks at Phoronix is Up..

As expected the HPC, 3D Fluid Dynamics and Deep Learning Performance is Off the Charts

View attachment 244881

View attachment 244882

View attachment 244884

I'm actually pretty surprised at how well the improvement for zstd was given how lackluster the WinRAR and 7z performance was on the TPU review. This could be further evidence that the CPU scheduler in the OS can directly impact hit ratios and evictions from cache due to how tasks are scheduled.

 

[nicamarvin]

I'm actually pretty surprised at how well the improvement for zstd was given how lackluster the WinRAR and 7z performance was on the TPU review. This could be further evidence that the CPU scheduler in the OS can directly impact hit ratios and evictions from cache due to how tasks are scheduled.

Would be nice if there is any Performance boost on Windows Games running on Linux + Proton/Wine(SteamOS) vs Windows 10/11 with 3D V-Cache.

What 5950X3D Might have been

Intel Core i9-12900K vs. AMD EPYC 7373X vs. AMD Ryzen 9 5950X Benchmarks - OpenBenchmarking.org

openbenchmarking.org

 

[Mussels]

In a review of the new "GAMING focussed AMD RYZEN R7 5800X3D " why would you choose to test 38 applications and only 10 games?
This defies common sense and logic !
It should have been 38 games tested and only 10 applications !
Maybe I missed something ? Confused ?

because that's what all previous CPU's were tested with, if they're not all tested the same way then you're not a reviewer - you're a shill


All parts must be tested equally, and then your conclusion is drawn off the results
(In this case, the 5800x3D has weak multi threaded performance compared to other ryzen and intel chips - but that's not a negative, for a gaming chip)

 

[Deleted member 24505]

Most people buying the X3D are not going to be bothered about app performance, if they were they would buy a different chip

 

[lexluthermiester]

Most people buying the X3D are not going to be bothered about app performance, if they were they would buy a different chip

Not true. Some will buy it for gaming as there is clearly some benefit. Others will be buying it for certain types of compute as there is a very clear advantage over CPU's with less cache.

 

[GURU7OF9]

I'm actually pretty surprised at how well the improvement for zstd was given how lackluster the WinRAR and 7z performance was on the TPU review. This could be further evidence that the CPU scheduler in the OS can directly impact hit ratios and evictions from cache due to how tasks are scheduled.

Imagine how good they will be on AMD EPYC cpus with the 3D V cache AND 64 cores ! TOTAL BEASTS !

Most people buying the X3D are not going to be bothered about app performance, if they were they would buy a different chip

Not true. Some will buy it for gaming as there is clearly some benefit. Others will be buying it for certain types of compute as there is a very clear advantage over CPU's with less cache.

TIGGER IS RIGHT! if they have done any sort of research for mainly app usage then they wouldnt buy the 5800X3D. If they did then they are just stupid !

because that's what all previous CPU's were tested with, if they're not all tested the same way then you're not a reviewer - you're a shill


All parts must be tested equally, and then your conclusion is drawn off the results
(In this case, the 5800x3D has weak multi threaded performance compared to other ryzen and intel chips - but that's not a negative, for a gaming chip)

I understand what you are saying but this is a GAMING orientated cpu . I would have thought that to keep it the same but still focus on the game performance, testing extra games would have been logical .
The bottom line is it has this nice shiny new 3D Vertical cache and all everyone wants to know is will it be usefull for more than only a few games!
Everyone already knows the performance of it in apps is going to be pretty much the same as the std 5800X +/- 5%, which is minimal difference !
So 38 apps have been basically retested for trivial results! I suppose just confirming it, but seems like a waste to me !

 

[Dr. Dro]

Truth is, "gaming" processors don't exist. The 5800X3D is marketed towards desktop workloads (which includes gaming) because of its relatively low core count (really, it is just one CCX/CCD) and AM4's relatively pedestrian I/O. Even with a single compute die, this processor, for the price it is sold will have a lot of buyers outside of the gaming spectrum, that's for sure.

 

[lexluthermiester]

TIGGER IS RIGHT! if they have done any sort of research for mainly app usage then they wouldnt buy the 5800X3D.

Tigger was making a specific point, which you failed to understand the context of. I was making an addendum to his point. You failed to understand that too. Context is not one of your strong suits it would seem.

If they did then they are just stupid !

Look in a mirror and say that again. Anyone who READS the benchmarks and reviews will know that the X3D has an advantage in certain area's over non-X3D parts. IF those areas of advantage are important to a user than the X3D is the CPU to buy.

I suppose just confirming it, but seems like a waste to me !

Then don't buy it.

 

[Dr. Dro]

Thought I'd share, Phoronix has tested the 5800X3D under Linux, where it somewhat surprisingly "sucks" at games, but excels at pretty much a wide range of other... business-ey things

AMD Ryzen 7 5800X3D On Linux: Not For Gaming, But Very Exciting For Other Workloads - Phoronix

www.phoronix.com

 

[btk2k2]

Some interesting results in the below writeup. Memory overclocking is far less important on Zen 3D than regular Zen. Also some fantastic improvements in Stellaris. I like how for the 1st metric he notes that as each day goes by the simulation gets more complex so comparing days in a fixed time does somewhat penalise faster parts. He then does time to day X comparison as well which I have attached below but it is well worth actually giving it a read.

Writeup here.

 

[GURU7OF9]

Tigger was making a specific point, which you failed to understand the context of. I was making an addendum to his point. You failed to understand that too. Context is not one of your strong suits it would seem.

Look in a mirror and say that again. Anyone who READS the benchmarks and reviews will know that the X3D has an advantage in certain area's over non-X3D parts. IF those areas of advantage are important to a user than the X3D is the CPU to buy.

Then don't buy it.

Well yes you are correct for very specific apps it may well be very good but it would be a very small amount ! But if it is what you need, then go for it!
But you have taken my comments out of text and you should take a good look in the mirror yourself!

Firstly my point agreeing with Tigger is a generalisation for 95 % of apps there are far better cpus (Intel and AMD) to use than the 5800X3D, except for some very small specific use cases.
Secondly they did the whole "testing procedure on a gaming orientated cpu, heavily focussing on application performance", with roughly only 25% of that amount on gaming, which is the cpus main focus!
(38 apps and only 10games)
Well that kind of speaks for itself.

Mussels went onto inform me it is because they are keeping it the same for review consistancy which is fair enough, but as i replied to him i would have thought they could have put in more/extra gaming benchmarks in the review to satisfy the masses, myself included, wanting to know how good the extra 3d cache helps for gaming!

You seem to think i have a problem with the 5800X3D, but i dont, its a neat bit of kit that i may well buy myself in the future!
Certainly thinking about it, but the high price is a bit rich at the minute! Hopefully it will drop before they run out of them!

 

[Deleted member 24505]

For a general use PC, there are better, for a mainly gaming rig, there is this was kind of my point.

 

[lexluthermiester]

Well yes you are correct for a wide range of specific apps

Fixed that for you..

 

[InVasMani]

Some interesting results in the below writeup. Memory overclocking is far less important on Zen 3D than regular Zen. Also some fantastic improvements in Stellaris. I like how for the 1st metric he notes that as each day goes by the simulation gets more complex so comparing days in a fixed time does somewhat penalise faster parts. He then does time to day X comparison as well which I have attached below but it is well worth actually giving it a read.

Writeup here.

I found this chart rather interesting overall. What would be interesting is seeing the 5800X3D tested at 3200MT/s and 3733MT/s and comparing temperatures in fact throwing in 2133MT/s would be nice as well and adding that to the performance comparisons. If you reduce the memory frequency MT/s it makes it easier to push BLCK higher. I don't know if pushing BLCK helps make pushing infinity fabric or not on the other hand. I do like the implications of this though on the performance impact side 3200MT/s isn't too far off from 3733MT/s memory so that seems to imply the CPU would handle 2x32GB DIMM's with slower latency timings and 4 DIMM's more reasonably by extension since reducing the CL to compensate for stability won't readily impact performance too aggressively.

 

[SL2]

What 5950X3D Might have been

Just no.

You'd better check the specs of the 7373X before posting anything like this. It takes 8 chiplets to get that amount of L3 cache, while AM4 only has room for 2 chiplets.

 

[lexluthermiester]

Just no.

Why not?

 

[SL2]

Why not?

See second line.