While not directly comparable, it gives a good idea of what the new RTX cards are capable of.

Folding@Home under Linux, GTX 680 up to and including RTX 2080Ti.

---

Grant
Darwin NT

i would have liked to see the 2080 vs the 1080ti in that test, but im guessing the tester only has a 2080ti on hand. the 2080 has been more comparable to the 1080ti in performance (games) benchmarks so far, and closer in pricing.

also a Titan XP vs the 2080ti might be a little better, but the results probably wouldnt be too different since it's pretty close to a 1080ti already.

the next gen cards should be truly exciting :). Price is still too high in my opinion, and ive seen reports that the price will only go UP thanks to the tariffs.

---

Seti@Home classic workunits: 29,492 CPU time: 134,419 hours

While not directly comparable, it gives a good idea of what the new RTX cards are capable of.

Folding@Home under Linux, GTX 680 up to and including RTX 2080Ti.

Wondered why the FAHBenchmark tests weren't included in the first tests in the article a couple days ago. Looks like they simply forgot and nice to see them now. Again, my original suspicion that for compute workloads, the Turing cards may not be that bad if one has no interests in gaming on the same system.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

The RTX 2070 will be available from Oct 17th, $for 200 less than the RTX 2080. Which makes it more expensive than a GTX 1080 (although if it's inline with the other TRX cards it should out perform a GTX 1080).

---

Grant
Darwin NT

Ask Petri to post his first 30 lines of his Event Log to show how BOINC reports his Titan V?

Wrong way round. I'm trying to re-write BOINC so that it tells Petri the truth.

Yes. And No. I do not need any version of Boinc telling me the truth about anything. You do not either.

Volta 80 sm. Each sm has 64 threads that have access to the fpu simultaneously (5120). And all 80 have their own cache and shared variables and 65k registers.

Pulse find algorithm can use all 80 SMs at a time to run 32 warps of 32 threads when scanning for pulses. The memory subsystem slows things more than the fact that there are only 5000+ cores executing fp32 instructions of those 80 thousand it could issue. And the Volta and I guess Turing can do address calculations (integer math) at the same time it is executing fp ops.

When I reboot next time I'll grab the log and post it here if anyone is interested. I'm sorry in advance if forget to do that. Please ask if you absolutely need it.

My guess is that the Turing performance will fall in the middle of the 1080Ti and Volta using less power than the Volta (144W) and way less than the 1080Ti (234W).

"Let me see of my 1080Ti. It has the capability to compete with the best of them. The power draw is just amazing - biggest I've ever seen. The 1080Ti I have is a big and heavy space heater. That's what I think."

---

To overcome Heisenbergs:
"You can't always get what you want / but if you try sometimes you just might find / you get what you need." -- Rolling Stones

I think that the reviews published and linked from this thread since I posted the message you quoted - disagree with you.

Did anyone knows if there is any RTX2080 actualy crunching SETI? So we could follow the performance in the real world.

If anyone could provide me with an access via terminal window (rlogin) to a machine with a RTX-2xxx/Ti and let me do the developing remotely ....

---

To overcome Heisenbergs:
"You can't always get what you want / but if you try sometimes you just might find / you get what you need." -- Rolling Stones

If the letters are RTX, the number will start with a 2.

My guess is that the Turing performance will fall in the middle of the 1080Ti and Volta using less power than the Volta (144W) and way less than the 1080Ti (234W).

From the Folding@Home review posted by Keith, two posts before yours:

The AC system power consumption when running FAHBench was 214 Watts on average with a peak of 319 Watts, compared to 196 Watts with the GTX 1080 Ti and a peak of 260 Watts.

That's for a RTX 2080 Ti.

If the letters are RTX, the number will start with a 2.

Rumors are the 2060 won't have the ray tracing cores, so probably GTX for this one and lower tier

EDIT: woops i've read your post backwards, your sentence is valid =P

---

I think that the reviews published and linked from this thread since I posted the message you quoted - disagree with you.

They can and they have the right to do so.

How do they disagree?
Did any of those reviews run Seti special app? or SoG?

Those 5120 cores that my Volta has - they do twice the job that my GTX1080 with 2560 cores. And my Volta is running at lesser clock speed 1552MHz vs 1923MHz.
The RTX has 34 sm and 128@time or 68 sm and 64@time - about 4352 cores. Both configurations can and will be used.

My power readings are from my daily driver and both the performance and the power draw can be estimated for the Turing being a derivative or a successor of the Volta architecture. I do not use RTX nor do I use tensor cores. Cuda cores, SMs and the cache/mem subsystem count.

I'll admit I'm wrong if proven so whenever I have a chance to compile and test with one [RTX] . I'm not going to buy one though until I've destroyed a total of four 1080GTX cards. (two still at the shelf).

---

To overcome Heisenbergs:
"You can't always get what you want / but if you try sometimes you just might find / you get what you need." -- Rolling Stones

If the letters are RTX, the number will start with a 2.

I'm sorry if you are having a bad day. My typo was a 1 instead of a 2 after the RTX.

---

To overcome Heisenbergs:
"You can't always get what you want / but if you try sometimes you just might find / you get what you need." -- Rolling Stones

Does the gddr6 memory on the new nvidia cards improve Seti performance compared to gddr5?

Does the gddr6 memory on the new nvidia cards improve Seti performance compared to gddr5?

that will likely depend on which app you are using.

---

Seti@Home classic workunits: 29,492 CPU time: 134,419 hours

Will it help the stock boinc app?

I would say yes, but it will be hard to quantify the speed improvement due to the memory change alone, since you will see speed improvements due to the GPU core changes as well.

---

Seti@Home classic workunits: 29,492 CPU time: 134,419 hours

I am trying to decide whether to buy a gtx 1070 or 1070ti or wait and get a 2060. I am planning to wait until it's released and see the reviews, and hopefully the gtx 1070ti prices drop more too.

i would say go for the 1070ti, and use the CUDA special app under linux.

you'll get great performance.

---

Seti@Home classic workunits: 29,492 CPU time: 134,419 hours

I am trying to decide whether to buy a gtx 1070 or 1070ti or wait and get a 2060. I am planning to wait until it's released and see the reviews, and hopefully the gtx 1070ti prices drop more too.

As far as I know, there haven't been any leaks or rumors about what architecture the lesser Turing cards will use. Assume an even more cut down version of the chip like the RTX 2070. Been some comments that the 2060, 2050 et al won't even use the RTX naming because there won't be any RT or Tensor cores in the chip and they will continue to use the GTX naming scheme.

So the only benefit in performance compared to 1000 series cards will be the use of the GDDR6 memory and the SM core improvements.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

If anyone out there has a P100, Titan V, or even RTX, it would be interesting to know whether the current BOINC versions (anything later than 2014) give a different peak flops value from SIV, and (in due course) whether BOINC v7.14 converges them again.

Hi Richard,
I have a pair of RTX2080's and have switched to Boinc 7.14. Is there anything that you want me to look for, or keep an eye on?
The new version 7.14 seems to be running OK, except for the reported memory in the CUDA section. Normal, I guess.

CUDA: NVIDIA GPU 0: GeForce RTX 2080 (driver version 411.70, CUDA version 10.0, compute capability 7.5, 4096MB, 3036MB available, 10598 GFLOPS peak)
CUDA: NVIDIA GPU 1: GeForce RTX 2080 (driver version 411.70, CUDA version 10.0, compute capability 7.5, 4096MB, 3036MB available, 10598 GFLOPS peak)
OpenCL: NVIDIA GPU 0: GeForce RTX 2080 (driver version 411.70, device version OpenCL 1.2 CUDA, 8192MB, 3036MB available, 10598 GFLOPS peak)
OpenCL: NVIDIA GPU 1: GeForce RTX 2080 (driver version 411.70, device version OpenCL 1.2 CUDA, 8192MB, 3036MB available, 10598 GFLOPS peak)

I have the latest version of SIV, but can't find the GFLOPS.
Let me know if I can help.

---

Bruce