PPL - Performance Per Liter, do you SFF?

petricor

Airflow Optimizer
Bronze Supporter
May 12, 2018
346
1,845
Just came across this and quickly benchmarked my current build - as-is right now, with no optimisation or overclocking whatsoever (running rock stable), incl overhead for remote access as running headless...

score:

2489451 - petricor - i7 8700k - GTX 1080 - (3806.5*3270)/5.0
validation link: https://valid.x86.fr/vtuftp

benchmarks:



build (slightly ahead of my log.. am a few posts behind!):
 

rfarmer

Spatial Philosopher
Jul 7, 2017
2,669
2,793
Just came across this and quickly benchmarked my current build - as-is right now, with no optimisation or overclocking whatsoever (running rock stable), incl overhead for remote access as running headless...

score:

2489451 - petricor - i7 8700k - GTX 1080 - (3806.5*3270)/5.0
validation link: https://valid.x86.fr/vtuftp

benchmarks:



build (slightly ahead of my log.. am a few posts behind!):

Nice score. I have been waiting for someone with a Skyreach to post their scores, figured it would do really well. Especially with a brickless build.
 
  • Like
Reactions: petricor

Duality92

Airflow Optimizer
Original poster
Apr 12, 2018
307
330
Just came across this and quickly benchmarked my current build - as-is right now, with no optimisation or overclocking whatsoever (running rock stable), incl overhead for remote access as running headless...

What power supplies are you running? I see the HDPlex 160, but that wouldn't do for the 1080 too?
 
  • Like
Reactions: petricor

QuantumBraced

Master of Cramming
Mar 9, 2017
507
358
I haven't read the thread, but I imagine the highest possible PPL would either be the new Deskmini without support for 2.5" drives with an 8700K and a 1080, a CustomMod Mini with a 7980XE and a 1080, or an S4M with a 7980XE and a 1080 Ti. Probably the CustomMod, if you can get one of the 3.2L versions that fits a 1080.
 

petricor

Airflow Optimizer
Bronze Supporter
May 12, 2018
346
1,845
What power supplies are you running? I see the HDPlex 160, but that wouldn't do for the 1080 too?
I use a MeanWell 200/12V for the 1080 (the long silver thing below the main board), and the HDPLEX 160 to power board and cpu - with a load switch between them. For details see my build log here
 

InfernoZeus

SFF Lingo Aficionado
May 23, 2017
98
53
edit: rewritten to be more clear.

A heatsink isn't a simple box, it's a stack of fins. Adding up the surface area of a stack of fins is almost exactly how you learn to calculate volume at the start of calculus. Heatsink surface area scales linearly with volume, not external area.

So if all the extra stuff takes up 25mm, and your case is 100m thick, your heatsink can use 75mm (75%) of the thickness, but if your case is 50mm thick, you only have 25mm (50%) of the room for it. So reducing your case thickness by 2x reduced your max heatsink size by 3x!
(Note: these are made up numbers and probably exaggerate the issue.)

So it's actually the reverse of your claim, thermals get worse at a greater than 1:1 ratio as you shrink the case thickness.

I was assuming that with a thinner case, it would also be wider. I.e. two cases with the same volume, one is almost a cube, the other a long thin rectangle.

Regardless, your point about surface area of the heatsink being tied more closely to the volume than the external surface area of the bounding box is a good one. Reasonably straightforward once you consider the density of the cooling fins.
 

RimsOnAToaster

Minimal Tinkerer
Jul 22, 2018
3
3
Updated everything, but I can't accept this methodology because if I compare my i5 7300U, which is an ES sample, so technically the way it's programmed acts like an i7 7500U, but your CPU-Z score is almost 800 points (80% more) than my CPU. The CPU-Z database has the first, initial benchmark, which was very Intel sided, so they changed it in like 1.83, since then, the benchmark scores have changed considerably. A stock i7 4790K scores 2300 for example, you're not 500 points off that CPU. I'd estimate your CPU score somewhere between 700 and 800.

I hope you understand and I think if you'd like to participate, you'd need to find yourself a an old SSD/HDD that you could do these benches.

Yeah, I had a feeling that score was a way too high, but I didn't have a point of reference to work with. I'll see if I can get a bootcamp instance rocking and try it again the right way.
 

Damascus

Master of Cramming
Feb 27, 2018
553
395
@Duality92 LMK if I missed anything, I'm tired as hell right now so something may have slipped by me

Here we go-

6950x + 1080 ti in a DSE breathe

4̶8̶4̶4̶ ̶x̶ ̶4̶4̶4̶3̶ ̶/̶ ̶7̶.̶3̶1̶ ̶=̶ ̶2̶,̶9̶4̶4̶,̶1̶7̶1̶ ̶(̶F̶o̶r̶g̶o̶t̶ ̶t̶h̶e̶ ̶e̶x̶t̶r̶a̶ ̶.̶0̶1̶ ̶i̶n̶ ̶m̶y̶ ̶i̶n̶i̶t̶i̶a̶l̶ ̶c̶a̶l̶c̶u̶l̶a̶t̶i̶o̶n̶)̶

edit - my CPU score goes up 200 points when heaven isn't on, updating to reflect that

5053 x 4443 / 7.31 = 3,075,408

Validation
https://valid.x86.fr/l19vj3

Case
Length: 183.1mm
Width: 115.1mm
Height: 346.9mm
Volume: 7.31L

 
Last edited:

Damascus

Master of Cramming
Feb 27, 2018
553
395
Nice job on the 3 million. You are right that 7980xe is brutal, made the score on my 8700k look mighty small. ;)
It really is amazing, and the 11L case was just big enough to cool it while still giving a s.all enough volume to be competitive.

I'm percolating an idea right now, an s4m with my cpu + gtx 1180 mini, with one internal hdplex 400w and a second sitting outside, of maybe I can use an additional, smaller brick internally for the cpu. I can re-use my cooler and the GPU score would be higher but the volume would be way smaller, to the point where I think a 6 million point score is feasible
 

n4ru

Cable Smoosher
Nov 7, 2016
10
12
  • Picture of actual PC
    My Battlestation
  • Volume calculated with exterior dimensions of case WITH power brick(s)
    Volume of NUC is 5.5 * 8.5 * 1.5 Cubic Inches = 1.14 Liters
    Volume of Power Brick is 3.8125 * 7.75 * 1 Cubic Inches = 0.48 Liters
  • CPU-Z / GPU-Z screenshots obligatory, see post #6
    CPU-Z / GPU-Z
  • CPU-Z must be run on "Benchmark" and validation link saved with your username
    2852 Multi-thread performance
  • GPU-Z must be on the first tab when doing screenshot with a dedicated GPU
    CPU-Z / GPU-Z
  • Heaven Benchmark on "Extreme" preset
    1426 Heaven Benchmark on Extreme Preset
  • PPL is calculated by (CPU-Z Multi Thread score * Heaven Benchmark score)/Build Volume rounded to nearest thousand.
    (2852 * 1426) / (1.14 + 0.48) = 2510464 Performance Per Liter Score
I think that just barely puts me into 3rd place :)
 
Last edited:

n4ru

Cable Smoosher
Nov 7, 2016
10
12
If I grab a GTX 1080 ITX and a meanwell power supply, I can score ~4.1m (after accounting for an extra 1.09 liters in volume and a 3200 GTX 1080 score after overclocking and docking some for running it via M.2) which would be an easy first place. Though at this point, I think a smaller NUC + an eGPU with a brick would score even higher.
 
Last edited: