Has anyone tested the 9900 temperatures with HT disabled?

[Machinus]

I would be very curious to see a comparison of the 9700 vs. the 9900 with HT disabled, specifically power draw and temperatures.

Has any site done a benchmark like this?

 

[The Stilt]

~6% higher power consumption for > 25% higher (average) MT performance.

 

[Indus]

Doesn't that make it the 9700k? 8 cores, 8 threads?

 

[Dave3000]

Doesn't that make it the 9700k? 8 cores, 8 threads?

Not exactly. The 9900k still has 4 MB more of L3 cache and a little higher turbo boost than the 9700k. Hyperthreading with 8 cores does nothing for gaming currently and the higher clocks only make it a few percent faster than the 9700k in the gaming benchmarks shown in recent reviews. What would be the point in buying a 9900k instead of a 9700k if you are going to disable hypertheading?

 

[Shmee]

As you just said, the 9900k has more cache, and if not OCing, it will turbo higher. If you are overclocking, I suspect the cache would not matter? or would it? Suppose both OC settle on 5GHz with 8c8t, how much faster is the 9900k due to extra cache?

 

[ZGR]

Not exactly. The 9900k still has 4 MB more of L3 cache and a little higher turbo boost than the 9700k. Hyperthreading with 8 cores does nothing for gaming currently and the higher clocks only make it a few percent faster than the 9700k in the gaming benchmarks shown in recent reviews. What would be the point in buying a 9900k instead of a 9700k if you are going to disable hypertheading?

I play Rainbow 6 Siege and 4c4t CPU's get around 50 FPS lower than 4c8t CPU's in Terrorist Hunt (max CPU load).

 

[CHADBOGA]

I play Rainbow 6 Siege and 4c4t CPU's get around 50 FPS lower than 4c8t CPU's in Terrorist Hunt (max CPU load).

He does say "Hyperthreading with 8 cores does nothing for gaming currently".

When you only have 4 cores, it is not so surprising if Hyperthreading helps in certain games.

 

[ZGR]

He does say "Hyperthreading with 8 cores does nothing for gaming currently".

When you only have 4 cores, it is not so surprising if Hyperthreading helps in certain games.

I have tested R6 on a 14c xeon, and HT didn't add such a considerable boost; but was still there. Clockspeed was also much lower, memory speed was much lower, and I was rather unimpressed with performance as a whole. I would love to see more R6 Terrorist Hunt CPU benchmarks. No matter how many cores you throw at it, it will report near 100% usage on each core. It isn't actually 100% usage, but it is pretty funny to see.

Keep in mind, the only mode on R6 that loads in enough terrorist at spawn in Disarm Bomb, and barely anyone plays that.

 

[RichUK]

Judging by the OP's avatar, this could well be a troll post.

 

[Dave3000]

I play Rainbow 6 Siege and 4c4t CPU's get around 50 FPS lower than 4c8t CPU's in Terrorist Hunt (max CPU load).

Are you sure it's not because of the combination of a larger L3 cache, faster clock speeds, and hyperthreading on the 4-core CPU, and not just because of hyperthreading accounting for the increased performance in the game?

 

[ZGR]

Are you sure it's not because of the combination of a larger L3 cache, faster clock speeds, and hyperthreading on the 4-core CPU, and not just because of hyperthreading accounting for the increased performance in the game?

Absolutely positive. I have tested this game on a wide variety of CPUs, some where the i5 has larger L3 than the i7, and where the i5 is clocked significantly higher.

I have also tested with hyperthreading disabled. The highest core CPU I used for R6 was a 14c/28t Xeon. Disabling HT has a negligible performance drop, but still there. I think having 8 cores without HT would be perfectly fine.