Discussion Ryzen 9000X3D series review thread

[Indus]

I have a 9800X3D and I noticed in OCCT in the normal CPU and the Linepack 2021 stability tests that my CPU will stay at slightly over 4.7GHz then after 30 seconds or so drop a little below its 4.7GHz base frequency as it gets closer to its 95C Tjmax temperature and it gets closer to 4.6GHz than 4.7GHz about a minute of staying at the 95C Tjmax in these two tests. The Linepack 2019 stability test, my 9800X3D stays between 5 to 5.2 GHz even at the 95C Tjmax temperature and the same with Cinebench R20. I use a Noctura U12A CPU cooler to cool my 9800X3D. Should my CPU guarantee it's 4.7GHz base frequency, no matter what, since it's advertised with a 4.7GHz base frequency? I thought base clocks were guaranteed minimum clocks by the manufacturer of the CPU no matter what. During gaming, it is usually running at 5.2-5.25GHz.

This is not an AMD thing.. I have a few Intel CPU's and they all drop down to below base frequency when idling. My 12600k idles at 400mhz when I'm not using it that much..

Check CPUID. The bus speed never drops but the multiplier should drop depending on idle or use or gaming states.




You should be thankful it drops below the advertised speed because it saves you energy and electricity costs.. as people actually have run it at full speed all the time by changing it in bios settings and it then becomes an major drain on electricity.

Mostly a CPU will look at states of use and priority:
1. Is it being used? No - It'll idle.
2. Yes? If being used, how high can it go without overheating?
3. If it gets too hot, it'll throttle back.

 

[igor_kavinski]

Back in the time CPUs did not have a turbo boost feature and only a base clock, would they throttle below their advertised frequency if they started to overheat or reach a specific temperature or would they just continue to stay at their advertised frequency and eventually shutdown the system or kill the CPU due to overheating or would they throttle as well?

A CPU will maintain its base clock as long as your cooling solution is able to dissipate the heat properly. If not, it will throttle to keep the CPU from overheating and causing heat related damage. Regarding 7800X3D, it is prevented from going above 89C. If you want a fair comparison, reduce the 9800X3D's max temp to 89C from its default 95C.

By the way, official pages for both CPUs mention liquid cooler for optimal performance.

 

[Dave3000]

A CPU will maintain its base clock as long as your cooling solution is able to dissipate the heat properly. If not, it will throttle to keep the CPU from overheating and causing heat related damage. Regarding 7800X3D, it is prevented from going above 89C. If you want a fair comparison, reduce the 9800X3D's max temp to 89C from its default 95C.

I just set the Tjmax to 89C in the BIOS and the CPU frequency start at just over 4.7GHz and slowly dropped to just under 4.6GHz after 4 1/2 minutes in the normal CPU test in OCCT.

 

[igor_kavinski]

I just set the Tjmax to 89C in the BIOS and the CPU frequency start at just over 4.7GHz and slowly dropped to just under 4.6GHz after 4 1/2 minutes in the normal CPU test in OCCT.

There is the slight possibility that you got a bad sample but we really need someone else to test this with their 9800X3D.

 

[Dave3000]

Also I tested my 7800X3D and on the same test in OCCT it reached 77C and was between between upper 4.2GHz to lower 4.3GHz, which is still above its base clock. Why didn't my 7800X3D go faster if it there is PPT and Tjmax headroom to do so as it was well under the 89C Tjmax for the 7800X3D the whole time I tested it? My 9800X3d was most of the time around 138W in this test and my 7800X3D was most of the time around 74W in this test. Shouldn't my 7800X3D go as fast as it can in this test until it reached 89C Tjmax and is still within the default PPT range?

Do you have a 7800X3D? Another thing weird I noticed is that in OCCT latency/bandwidth benchmark, my 7800X3D shows much slower L3 cache bandwidth than my 9800X3D. My 7800X3D gets around 154MB/s read and 75MB/s write for L3 cache and my 9800X3D gets around 541MB/s read and 298MB/s write for L3 cache in this benchmark. In AIDA64, my 7800X3D is a little slower than my 9800X3D in the L3 cache benchmark, so I wonder if something is wrong with my 7800X3D that would cause the much lower L3 cache performance in OCCT than my 9800X3D. Maybe I also have a bad sample of my 7800X3D? Anyone else here with a 7800X3D have this weird issue in the L3 cache test in the latency/bandwidth benchmark in OCCT or should it be close to what the 9800X3D gets for L3 cache performance just like in AIDA64?

 

[igor_kavinski]

Why didn't my 7800X3D go faster if it had plently of TDP and Tjmax headroom to do so as it was well under the 89C Tjmax for the 7800X3D the whole time I tested it? My 9800X3d was most of the time around 138W in this test and my 7800X3D was most of the time around 74W in this test. Shouldn't my 7800X3D go as fast as it can in this test until it reached 89C Tjmax and is still within the default TDP range?

It is limited by the V-cache lying on top of the CCD. If it goes any faster, the cache will crash and burn.

9800X3D has the V-cache at the bottom of the CCD.

Do you have a 7800X3D? Another thing weird I noticed is that in OCCT latency/bandwidth benchmark, my 7800X3D shows much slower L3 cache bandwidth than my 9800X3D. My 7800X3D gets around 157MB/s read and 74MB/s write for L3 cache and my 9800X3D gets around 480MB/s read and 340MB/s write for L3 cache in this benchmark. In AIDA64, my 7800X3D is a little slower than my 9800X3D in the L3 cache benchmark, so I wonder if something is wrong with my 7800X3D that would cause the much lower L3 cache performance in OCCT than my 9800X3D. Maybe I also have a bad sample of my 7800X3D?

9950X3D. I can't test it right now (not hooked up) and hooking it up is a bit of an effort since it's normal configuration is backside ports towards the floor which makes plugging stuff in it challenging. Had to do it that way coz it's a small mATX case and I wanted the 240mm AIO to be on top.

7800X3D is 2nd generation V-cache. That's the best AMD could do at the time. I wouldn't care that much about the lower L3 performance. It's hard to be performant for that cache when it's literally being cooked by the CCD underneath it.

 

[Dave3000]

It is limited by the V-cache lying on top of the CCD. If it goes any faster, the cache will crash and burn.

9800X3D has the V-cache at the bottom of the CCD.


9950X3D. I can't test it right now (not hooked up) and hooking it up is a bit of an effort since it's normal configuration is backside ports towards the floor which makes plugging stuff in it challenging. Had to do it that way coz it's a small mATX case and I wanted the 240mm AIO to be on top.

7800X3D is 2nd generation V-cache. That's the best AMD could do at the time. I wouldn't care that much about the lower L3 performance. It's hard to be performant for that cache when it's literally being cooked by the CCD underneath it.

My concern is not with Ryzen 9000X3D series in this benchmark for L3 cache performance, so don't bother doing that.

 

[StefanR5R]

Why didn't my 7800X3D go faster if it there is PPT and Tjmax headroom

Besides limits to power and temperature, there are also limits to Voltages and currents.
AFAIU, the new X3D stacking of the 9000X3D series does not only improve thermal conduction from cores to cooler, but also allows for higher core Voltages compared to 5000X3D/7000X3D.