- Joined
- Oct 6, 2004
- Messages
- 58,413 (8.18/day)
- Location
- Oystralia
System Name | Rainbow Sparkles (Power efficient, <350W gaming load) |
---|---|
Processor | Ryzen R7 5800x3D (Undervolted, 4.45GHz all core) |
Motherboard | Asus x570-F (BIOS Modded) |
Cooling | Alphacool Apex UV - Alphacool Eisblock XPX Aurora + EK Quantum ARGB 3090 w/ active backplate |
Memory | 2x32GB DDR4 3600 Corsair Vengeance RGB @3866 C18-22-22-22-42 TRFC704 (1.4V Hynix MJR - SoC 1.15V) |
Video Card(s) | Galax RTX 3090 SG 24GB: Underclocked to 1700Mhz 0.750v (375W down to 250W)) |
Storage | 2TB WD SN850 NVME + 1TB Sasmsung 970 Pro NVME + 1TB Intel 6000P NVME USB 3.2 |
Display(s) | Phillips 32 32M1N5800A (4k144), LG 32" (4K60) | Gigabyte G32QC (2k165) | Phillips 328m6fjrmb (2K144) |
Case | Fractal Design R6 |
Audio Device(s) | Logitech G560 | Corsair Void pro RGB |Blue Yeti mic |
Power Supply | Fractal Ion+ 2 860W (Platinum) (This thing is God-tier. Silent and TINY) |
Mouse | Logitech G Pro wireless + Steelseries Prisma XL |
Keyboard | Razer Huntsman TE ( Sexy white keycaps) |
VR HMD | Oculus Rift S + Quest 2 |
Software | Windows 11 pro x64 (Yes, it's genuinely a good OS) OpenRGB - ditch the branded bloatware! |
Benchmark Scores | Nyooom. |
This is half project log, half proving some points about power supplies that i see people comment on without any factual knowledge.
What i'm doing is swapping my PSU from a 750W to a 1000W, and recording the wattages in various tests. I'm expecting a slight wattage decrease especially in the higher wattage tests, not sure how it will vary in the windows idle test.
I am testing:
Off
S3 sleep
BIOS
Idle in windows
Extreme load in windows (synthetic tests)
Load in games with Vsync on and off
test done with following settings:
Test results:
Conclusions from tests
Pictures, since everyone loves pictures!
What i'm doing is swapping my PSU from a 750W to a 1000W, and recording the wattages in various tests. I'm expecting a slight wattage decrease especially in the higher wattage tests, not sure how it will vary in the windows idle test.
I am testing:
Off
S3 sleep
BIOS
Idle in windows
Extreme load in windows (synthetic tests)
Load in games with Vsync on and off
test done with following settings:
Monitor NOT included - measuring tower only.
3.9Ghz 1090TBE @ 1.375v
Ram @ 1600Mhz 1.65v
NB + HT @ 2.4GHz (stock volts)
5870@stock (no OC or voltage adjusts at all)
1x samsung F1 500GB (OS drive)
2xmaxtor 250GB in raid 0 (games drive)
1x optical drive
~5 120mm fans (hardly relevant, low wattage fans)
3.9Ghz 1090TBE @ 1.375v
Ram @ 1600Mhz 1.65v
NB + HT @ 2.4GHz (stock volts)
5870@stock (no OC or voltage adjusts at all)
1x samsung F1 500GB (OS drive)
2xmaxtor 250GB in raid 0 (games drive)
1x optical drive
~5 120mm fans (hardly relevant, low wattage fans)
Test results:
Off
750 - 3W
1000 - 3W
'Idle' (BIOS)
750 - 245W
1000W - 245W - yup, identical
Idle (Win Desktop)
750 - 159W
1000 - 169W (higher - POSSIBLY due to a HDD idling on first PSU test)
standby (S3 sleep)
750 - 5W (note how its barely more than when its off)
1000 - 5W
Load (OCCT linpack + furmark peak) (this is maximum burn - no game or app will ever get this high, but you should make sure your system can handle it anyway)
750 - 462W
1000 - 448W - lower! this PSU's efficiency is in the higher wattages
and for a repeatable realistic game test...
Company of heroes, DX9 max: in game benchmark. peak value is recorded.
Main Menu:
750 - 202W (Vsync off: 267W)
1000 - 205W (Vsync off: 272W)
Benchmark (Vsync on):
750 - 291W - for about 2 seconds, most was in the 250-260W range
1000 - 300W for about 1 second, mostly 248W - slightly lower
Benchmark (Vsync off):
750 - 339 avg 320
1000 - 319W - 350W - better and worse at the same time, i call that 'near enough'
750 - 3W
1000 - 3W
'Idle' (BIOS)
750 - 245W
1000W - 245W - yup, identical
Idle (Win Desktop)
750 - 159W
1000 - 169W (higher - POSSIBLY due to a HDD idling on first PSU test)
standby (S3 sleep)
750 - 5W (note how its barely more than when its off)
1000 - 5W
Load (OCCT linpack + furmark peak) (this is maximum burn - no game or app will ever get this high, but you should make sure your system can handle it anyway)
750 - 462W
1000 - 448W - lower! this PSU's efficiency is in the higher wattages
and for a repeatable realistic game test...
Company of heroes, DX9 max: in game benchmark. peak value is recorded.
Main Menu:
750 - 202W (Vsync off: 267W)
1000 - 205W (Vsync off: 272W)
Benchmark (Vsync on):
750 - 291W - for about 2 seconds, most was in the 250-260W range
1000 - 300W for about 1 second, mostly 248W - slightly lower
Benchmark (Vsync off):
750 - 339 avg 320
1000 - 319W - 350W - better and worse at the same time, i call that 'near enough'
Conclusions from tests
A few things come to mind here, that arent obvious to most users.
1. idling in the BIOS consumes a fair bit of power, and outputs a fair amount of heat - almost as much as gaming! BIOS should NOT be considered idle, when watching temps!
2. Vsync drastically alters the power used, in high end systems. Ever wondered why one game causes you to overheat and crash, but not another? default Vsync settings could be why...
2.1 When people start contradicting common knowledge about certain cards being extremely hot (cough, fermi), this is probably why. reviewers test with Vsync off, these games having Vsync on will lower the power use and heat output drastically. I saved 40-70W turning Vsync on in CoH, making my temp controlled PSU fan much quieter.
3. S3 sleep mode is awesome. Barely using more power than a system that is off, yet you can resume whatever you were doing instantly.
4. The advice not to get an overkill PSU (a 1000W PSU for a 300W PC) is quite solid. you will save power and thus heat, getting a PSU that matches your requirements. Just remember that if you get a multi rail PSU that each rail is sufficient for what you'll have on them (EG, dont get a 700W with four 18A rails if you want to run a fermi - get one with two 35A rails instead)
4.1 since the HX1000 was a better quality PSU (more efficient design), moving out of its sweetspot is pretty much on par with the 750W anyway.
5. slightly contradicting above, this 1000W PSU has its fan spin up at a higher load/temperature, so its actually quieter than the 750W, despite using slightly more power. Thats worth considering at least.
1. idling in the BIOS consumes a fair bit of power, and outputs a fair amount of heat - almost as much as gaming! BIOS should NOT be considered idle, when watching temps!
2. Vsync drastically alters the power used, in high end systems. Ever wondered why one game causes you to overheat and crash, but not another? default Vsync settings could be why...
2.1 When people start contradicting common knowledge about certain cards being extremely hot (cough, fermi), this is probably why. reviewers test with Vsync off, these games having Vsync on will lower the power use and heat output drastically. I saved 40-70W turning Vsync on in CoH, making my temp controlled PSU fan much quieter.
3. S3 sleep mode is awesome. Barely using more power than a system that is off, yet you can resume whatever you were doing instantly.
4. The advice not to get an overkill PSU (a 1000W PSU for a 300W PC) is quite solid. you will save power and thus heat, getting a PSU that matches your requirements. Just remember that if you get a multi rail PSU that each rail is sufficient for what you'll have on them (EG, dont get a 700W with four 18A rails if you want to run a fermi - get one with two 35A rails instead)
4.1 since the HX1000 was a better quality PSU (more efficient design), moving out of its sweetspot is pretty much on par with the 750W anyway.
5. slightly contradicting above, this 1000W PSU has its fan spin up at a higher load/temperature, so its actually quieter than the 750W, despite using slightly more power. Thats worth considering at least.
Pictures, since everyone loves pictures!
just comparing the two PSU's and before/after of the inside of my case. irrelevant to tests.
Last edited: