Thanks Shaggie76 You are certainly putting some effort into your data collection :-)

---

Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?

@Shaggie76: Is it possible to get some stats about SoG vs Cuda 5.0 on GTX7xx, GTX9xx and GTX10x0?

---

@Shaggie76: Is it possible to get some stats about SoG vs Cuda 5.0 on GTX7xx, GTX9xx and GTX10x0?

I was working on that but ran out of time last night. Of course the instant I post it people will start clamoring about it being meaningless because if you're going to run CUDA you need to run multiple tasks to get the same throughput...

They do have a point. It's like comparing apples and oranges.

Ideally you want same number of work units on both types of GPUs to see how they do.

We know running multiple cuda decreases the overall time compared to a single cuda work unit

The same can be said for SoG.

So single instances on either type of App doesn't reflect what they could actually be doing.

Then there is the commandlines that increase the efficency of the work.

But that being said, it makes for some interesting reading.

Carry on Shaggie ;)

@Shaggie76: Is it possible to get some stats about SoG vs Cuda 5.0 on GTX7xx, GTX9xx and GTX10x0?

I was working on that but ran out of time last night. Of course the instant I post it people will start clamoring about it being meaningless because if you're going to run CUDA you need to run multiple tasks to get the same throughput...

That is not necessarily so true. I run One at a time. There are some other too (currently Linux and Mac) who do run CUDA (special) who are running one at a time too. We are under the label 'Anonymous platform'.

I guess that You(Shaggie76) have the means and the ability to run a comparison for reference of what to expect when deciding what to purchase in the future.

This is by no means to ment to be a war-ignitor between NV and AMD loving people. I know that the same techniques I have used can be applied to OpenCL thus contributing for the good of both camps. We need patience.

... and Think about the time to travel to our future nearest neighbor waving at us with a pulse, triplet, autocorr, ... or whatever we can understand as intelligent... unless we are just an other XtraTerrestialSetiSimulation in their terrarium/terraria.


And then another observation. The Fuji/Fury whatever big variance... Someone must know how to get the most of it with the current options.

Petri. (EDIT: double signature removed)

---

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 figured since we're having an Arecibo party this weekend it would be fun to do a GPU scan. One nice thing about this is it greatly reduced the variance in the data slightly.

Not enough Fiji hosts in action today to count but a whole lot of 1060's -- they're all 6GB parts so far, too, so I'm not treating them distinctly.

My wallet just groaned, seeing the 10 series hanging with 750ti for efficiency.

---

"Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions.

My wallet just groaned, seeing the 10 series hanging with 750ti for efficiency.

You know it. I had just gotten a couple 980s, and I'm sitting here thinking the right thing to do is build a 750ti farm and hang it all external. Sheesh...

---

My wallet just groaned, seeing the 10 series hanging with 750ti for efficiency.

. . OK Jason now take a deep breath and look back at the credit per hour comparison :)

. . Feeling better again ? :)

. . I am kicking myself for buying 2 x GTX970s a couple of months ago. I should have waited (if I had known the 1060s were so close) slightly better output and MUCH better efficiency, and about the same price :(

. . Not to mention physically smaller and cooler running :)



.

Yeah, for me the 1070's are looking viable as lower power higher throughput than everything in the house. 1060's would probably temporarily trim output a little, but give more beer money, so increase development output. Am going to have to mull over priorities/budget versus costs, which is why I like this investigation a lot.

---

"Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions.

My wallet just groaned, seeing the 10 series hanging with 750ti for efficiency.

You know it. I had just gotten a couple 980s, and I'm sitting here thinking the right thing to do is build a 750ti farm and hang it all external. Sheesh...

. . Or the ASUS/MSI GTX950s that do NOT require external PCIe power and crunch full tilt using only about 65W. They are right up there too.

My wallet just groaned, seeing the 10 series hanging with 750ti for efficiency.

You know it. I had just gotten a couple 980s, and I'm sitting here thinking the right thing to do is build a 750ti farm and hang it all external. Sheesh...

Yeah.
The GTX 750Tis may not produce a lot of work compared to the others, but their power consumption in absolute terms is bugger all.
Compared to other cards of similar power ratings, their output is miles ahead.
They still are an amazing card in terms of work done for power used.
A GTX 1050Ti would make things very interesting.

---

Grant
Darwin NT

Yeah, for me the 1070's are looking viable as lower power higher throughput than everything in the house.

It's just a shame about their current pricing.

---

Grant
Darwin NT

Yeah, for me the 1070's are looking viable as lower power higher throughput than everything in the house.

It's just a shame about their current pricing.

Yep ... Australia tax not helping

---

"Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions.

My wallet just groaned, seeing the 10 series hanging with 750ti for efficiency.

You know it. I had just gotten a couple 980s, and I'm sitting here thinking the right thing to do is build a 750ti farm and hang it all external. Sheesh...

Soon.. very soon... I have most of the parts I need, but not the time. That is looking to change this week though, as I am calling in the cavalry to help me put this outdoor project to bed. I've finally had enough, it's not that I _can't_ do it, it's that there aren't enough me's, and hours in the day, to get everything I need to do done each day. So, hence the help soon, which will allow me to get on to more important things, like finishing that 750Ti super build I've been back-burnering for too long! lol

---

Yeah, for me the 1070's are looking viable as lower power higher throughput than everything in the house.

It's just a shame about their current pricing.

Yeah, I'm hoping something from AMD comes out that competes directly in that space, and helps drive down the cost of the 1070's in the 40-50% range in not too distant future (hey, a guy can dream, right?). ;-)

And still wondering if/when something like a 1050Ti will be built, and how it will compare to the 750?

---

Hi Shaggie,

Would it be possible to run a targeted scan of my (petri33) host and Mr Kevvys Linux hosts to get insight of Arecibo performance with special CUDA app. -=Vyper=- is running special too on his Linux.

All specials run one at a time.

My power Draw is between 120-170W, mostly 136W with Arecibo work load.


That is of interest since I wonder if productivity comes wit a cost.

---

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

You first, then Mr Kevvy's Linux Hosts, -= Vyper =- last.

C:\Scripts\SETI>aggregate.pl -anon 7475713 7985986 7939003 7961370 7957088 7945042 8018193 7944247 8053171
Host, API, Device, Credit, Seconds, Credit/Hour, Work Units
7475713, cpu, Intel Core i7-3930K @ 3.20GHz, 3886.42, 13523.6966666667, 1034.56268983653, 39
7475713, gpu, [4] NVIDIA GeForce GTX 1080, 166123.22, 185855.74, 3217.78381447891, 1961
7985986, cpu, Intel Core i7-5820K @ 3.30GHz, 33539.71, 66401.3341666667, 1818.38147554289, 456
7985986, gpu, [4] NVIDIA GeForce GTX 980, 126160.17, 393070.560000001, 1155.45822612611, 1544
7939003, cpu, Intel Core i7-3930K @ 3.20GHz, 32818.28, 61109.2008333333, 1933.35547493455, 395
7939003, gpu, [3] NVIDIA GeForce GTX 980, 125373.4, 525530.36, 858.835710271811, 1432
7961370, cpu, Intel Core 2 Quad Q6600 @ 2.40GHz, 6153.29, 46733.015, 474.008449914905, 72
7961370, gpu, [2] NVIDIA GeForce GTX 970, 83152.92, 296877.1, 1008.3314341187, 990
7957088, cpu, Intel Core 2 Quad @ 2.40GHz, 6647.92, 38441.1175, 622.575865542931, 80
7957088, gpu, NVIDIA GeForce GTX 970, 40557.12, 275369.12, 530.217883544821, 490
7945042, cpu, AMD FX(tm)-8350 Eight-Core Processor, 16051.1, 74594.60125, 774.639974364097, 216
7945042, gpu, NVIDIA GeForce GTX 970, 29634.02, 86945.7199999999, 1227.0008460451, 349
8018193, cpu, AMD Phenom(tm) 9500 Quad-Core Processor, 7523.62, 42741.99, 633.686732882582, 99
8018193, gpu, NVIDIA GeForce GTX 970, 38106.68, 185526.88, 739.42949938036, 456
7944247, cpu, AMD Phenom(tm) II X6 1045T Processor, 21680.56, 69026.7316666666, 1130.72159314897, 280
7944247, gpu, NVIDIA GeForce GTX 970, 24682.34, 177843.93, 499.631469007685, 311
8053171, gpu, [4] NVIDIA GeForce GTX 750 Ti, 93908.0399999999, 388460.45, 870.278928009272, 1060

You first, then Mr Kevvy's Linux Hosts, -= Vyper =- last.

C:\Scripts\SETI>aggregate.pl -anon 7475713 7985986 7939003 7961370 7957088 7945042 8018193 7944247 8053171
Host, API, Device, Credit, Seconds, Credit/Hour, Work Units

7475713, cpu, Intel Core i7-3930K @ 3.20GHz, 3886.42, 13523.6966666667, 1034.56268983653, 39
7475713, gpu, [4] NVIDIA GeForce GTX 1080, 166123.22, 185855.74, 3217.78381447891, 1961

7985986, cpu, Intel Core i7-5820K @ 3.30GHz, 33539.71, 66401.3341666667, 1818.38147554289, 456
7985986, gpu, [4] NVIDIA GeForce GTX 980, 126160.17, 393070.560000001, 1155.45822612611, 1544

7939003, cpu, Intel Core i7-3930K @ 3.20GHz, 32818.28, 61109.2008333333, 1933.35547493455, 395
7939003, gpu, [3] NVIDIA GeForce GTX 980, 125373.4, 525530.36, 858.835710271811, 1432

7961370, cpu, Intel Core 2 Quad Q6600 @ 2.40GHz, 6153.29, 46733.015, 474.008449914905, 72
7961370, gpu, [2] NVIDIA GeForce GTX 970, 83152.92, 296877.1, 1008.3314341187, 990

7957088, cpu, Intel Core 2 Quad @ 2.40GHz, 6647.92, 38441.1175, 622.575865542931, 80
7957088, gpu, NVIDIA GeForce GTX 970, 40557.12, 275369.12, 530.217883544821, 490

7945042, cpu, AMD FX(tm)-8350 Eight-Core Processor, 16051.1, 74594.60125, 774.639974364097, 216
7945042, gpu, NVIDIA GeForce GTX 970, 29634.02, 86945.7199999999, 1227.0008460451, 349

8018193, cpu, AMD Phenom(tm) 9500 Quad-Core Processor, 7523.62, 42741.99, 633.686732882582, 99
8018193, gpu, NVIDIA GeForce GTX 970, 38106.68, 185526.88, 739.42949938036, 456

7944247, cpu, AMD Phenom(tm) II X6 1045T Processor, 21680.56, 69026.7316666666, 1130.72159314897, 280
7944247, gpu, NVIDIA GeForce GTX 970, 24682.34, 177843.93, 499.631469007685, 311

8053171, gpu, [4] NVIDIA GeForce GTX 750 Ti, 93908.0399999999, 388460.45, 870.278928009272, 1060

Thank You! The gtx 750 Ti is putting out some serious numbers!

---

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 should add that the GPU data is per card -- I can't tell which card did which task (part of why I have to exclude multi-gpu setups from my scans).

Your quad 1080 rig is more like 12871 cr/h in GPU throughput -- pretty amazing.