Hi I have a pc with the following specs: MSI B350 Bazooka Motherboard, Ryzen 7 1700X, GTX 1070Ti, Thermaltake Smart 600W PSU, 1TB HDD, 160GB SSD, and 3 cooler master 140mm fans and want to know if I can run a 1080 Ti with my current power supply or should upgrade the psu as I want to eventually upgrade to either a GTX 1080 Ti or a Turing GTX 2080 in the summer.
600 Watt PSU enough for a 1080 Ti/ Future 2080?
- Thread starter jakhalil
- Start date
Solution
- Dec 26, 2012
- 30,756
- 4,559
- 142,790
You want to power $2k+ worth of pc on a $40 psu that's not a good psu to begin with? Recommended psu for any 1080ti is 650w. This might be a little overkill in some cases, but generally the better grade psus come in 550/650/750w sizes, there's almost no psus of 500/600/700w that I'd consider recommending.
Is it enough? Yes, barely, if you keep the OC to minimums, but high OC takes clean output ripples for stability, and that covers cpus, ram or gpus.
The TT Smart series are pretty much TT's version of the older Corsair CX. For a comparison.
Corsair RMx, Seasonic Focus Plus Gold, Evga G2/G3 650w would be solid investments.
Is it enough? Yes, barely, if you keep the OC to minimums, but high OC takes clean output ripples for stability, and that covers cpus, ram or gpus.
The TT Smart series are pretty much TT's version of the older Corsair CX. For a comparison.
Corsair RMx, Seasonic Focus Plus Gold, Evga G2/G3 650w would be solid investments.
quintnelson14
Commendable
- Oct 17, 2017
- 30
- 0
- 1,530
if you want to run it you will be able to (maybe). obviously I would never recommend it but I mean the 1080 ti could run with 600w PSU it would just sound like a jet and you may run into problems in the future because you are wearing out your PSU and windows may let you know you aren't getting enough power to the system. overclocking is a no no. my recommendation is to just get a 850-1000 psu and never be in this situation again, that way you could look at it as an investment (not really) you don't have to keep buying new one's for the future!
you can,
it won't be ideal but, you can definitely run it stock with no oc on a 600w psu without major issues.
as long as the brand are reputable, and your amp draw on 12v are well within spec.
best way to find out, is to get a wattage meter on your wall and connect just your pc to it, measure the power draw on your existing system,times that by 0.9 (that gives your power draw to match the output, 0.9 is the best case scenario for 80+bronze and silver,0.95 for 80+g,80+p, 1 for 80+t, for efficiency)
example: if you have 400w reading, you times 0.9 to get 360w, that's how much your psu is supplying to these complements,
i'm going to assume most of your power come from the 12V rails, it's rare that 3.3v and 5v are overloaded.
divide the 360w by the maximum amp rating on 12v rail (42 in your case, 12*42 = 500w) = 0.72, so your psu is running at 72% of capacity at most, as bit high, but doable.
here are the ranges.
load/12v*amp rating:
25% or below - overkill
25%~40% - still overkill, ideal if you plan to add future upgrades issue.
40%~60% - ideal (PSU runs most efficient at this point)
60%~75% - stretching it
75% or higher need upgrade.
it won't be ideal but, you can definitely run it stock with no oc on a 600w psu without major issues.
as long as the brand are reputable, and your amp draw on 12v are well within spec.
best way to find out, is to get a wattage meter on your wall and connect just your pc to it, measure the power draw on your existing system,times that by 0.9 (that gives your power draw to match the output, 0.9 is the best case scenario for 80+bronze and silver,0.95 for 80+g,80+p, 1 for 80+t, for efficiency)
example: if you have 400w reading, you times 0.9 to get 360w, that's how much your psu is supplying to these complements,
i'm going to assume most of your power come from the 12V rails, it's rare that 3.3v and 5v are overloaded.
divide the 360w by the maximum amp rating on 12v rail (42 in your case, 12*42 = 500w) = 0.72, so your psu is running at 72% of capacity at most, as bit high, but doable.
here are the ranges.
load/12v*amp rating:
25% or below - overkill
25%~40% - still overkill, ideal if you plan to add future upgrades issue.
40%~60% - ideal (PSU runs most efficient at this point)
60%~75% - stretching it
75% or higher need upgrade.
- Dec 26, 2012
- 30,756
- 4,559
- 142,790
That would be nice, mathematical breakdown. However, not all psus are built the same, to the same standards, with the same quality or quality of output.
https://youtu.be/f6snWfd1v7M
Take note of the psu printed and claimed wattage vrs the wattage at failure. For most, anything above the 40-60% range is not 'stretching it', it's meltdown, sometimes epically.
https://youtu.be/f6snWfd1v7M
Take note of the psu printed and claimed wattage vrs the wattage at failure. For most, anything above the 40-60% range is not 'stretching it', it's meltdown, sometimes epically.
go for any 80 plus certified psu, and double check on official list,
https://plugloadsolutions.com/80PlusPowerSupplies.aspx
some manufacture *claims* that they have 80 plus, but are not on the list
it means this model has been submitted for sampling and tested at various load point, you can also get a report regarding it's efficiency % at different load point, however, it does not tell anything regarding manufacture's quality assurance / control . but at least that's what it is designed to do and been tested against.
- Dec 26, 2012
- 30,756
- 4,559
- 142,790
It's physically impossible to mechanically cool below ambient temps. In basic, that means a fan blowing ambient air on a piece of electronics will not cool lower than the surrounding air it's in, you'd need a chemical process like evaporation to do so.
That said, every unit that's submitted to 80+ for testing and evaluation is tested at a regulated temp, it's the exact same temp for all units tested or the results would be non-reliable. Psus are tested at below 30°C ambient. This is done because many psus will loose power at above 30°C and therefore efficiency. Quality psus will handle @50°C for the most part. This is important, especially considering that most psus are already starting at a 30°C temp, just as most cpus, cases, gpus etc are running at 30°C+ just at idle. Even in 22°C ambients.
Now apply a good sized load, maybe a little dust on the filter, ambients above 22-23°C and low quality psus are at a severe disadvantage, loosing power, efficiency, output regulation, ripple control etc.
This leads to pc instability, possible damage to both the psu, cpu and pc in general.
80+ certification is nice, but in no way guarantees anything beyond what was available at the time of testing, in no way guarantees you get the same unit as tested (quite a few are specially built, cherry picked psus and not equitable to production standards).
That said, every unit that's submitted to 80+ for testing and evaluation is tested at a regulated temp, it's the exact same temp for all units tested or the results would be non-reliable. Psus are tested at below 30°C ambient. This is done because many psus will loose power at above 30°C and therefore efficiency. Quality psus will handle @50°C for the most part. This is important, especially considering that most psus are already starting at a 30°C temp, just as most cpus, cases, gpus etc are running at 30°C+ just at idle. Even in 22°C ambients.
Now apply a good sized load, maybe a little dust on the filter, ambients above 22-23°C and low quality psus are at a severe disadvantage, loosing power, efficiency, output regulation, ripple control etc.
This leads to pc instability, possible damage to both the psu, cpu and pc in general.
80+ certification is nice, but in no way guarantees anything beyond what was available at the time of testing, in no way guarantees you get the same unit as tested (quite a few are specially built, cherry picked psus and not equitable to production standards).
even though, there are many flaws with the 80 plus system, it still presents value to the consumer. consider it as minimum standards. Especially for consumer grade pc, i will go with a 80 plus psu vs a non 80 plus psu any day.
i mention nothing about the reliability/quality control aspect of the psu, only from performance point of view,
it's better to have some data and testing vs nothing at all.
I also don't think there is any thing wrong using it's values, as a rough estimate, as there is already some consideration of margin in the calculation.
- Dec 26, 2012
- 30,756
- 4,559
- 142,790
Not really. Personally speaking, I'd put an old corsair HX, non tested, non 80+ up against a thermaltake smart gold any day of the week.
But as said, it's a voluntary certification, it's not a standard by any means. It's abused frequently, and many cheap psus will throw the label on the box just to make it sound good. I've read several reviews of psus claiming certain 80+ levels, that when reviewed, failed. Sometimes miserably.
The unit should speak for itself, 80+ won't make it better, just sound better to salesmen.
But as said, it's a voluntary certification, it's not a standard by any means. It's abused frequently, and many cheap psus will throw the label on the box just to make it sound good. I've read several reviews of psus claiming certain 80+ levels, that when reviewed, failed. Sometimes miserably.
The unit should speak for itself, 80+ won't make it better, just sound better to salesmen.
- Oct 6, 2014
- 22,062
- 2,271
- 104,240
If your running your PSU at 75% under full load their nothing wrong with that, Full load is not really a real use load unless you do a lot of rendering that can use GPU acceleration and all your cores.
Most quality power supplies can output more than their rated capacity.
The load on the PSU does not hurt the PSU only HEAT.
2 or 3 points drop on the efficiency (again nothing that concerns me)
I usually go for the 66% (2/3) range under full load unless upgrades are planed, so we will have to agree to disagree their. Again 75% I would not be stressing over since for your average user your not going to be running it that high.
For the original poster 42A= 504 watts. (on the tell tell 12 volt rail)
For the processor 95 Watts (this can vary depending on what the motherboard defaults the Vcore to)
Some of the Asus GTX 1080ti's using a torture test can peak out at 285 watts (not a realistic load but could happen)
http://www.guru3d.com/articles_pages/asus_rog_strix_geforce_gtx_1080_ti_review,7.html
For the rest of the PC I always figure 60 watts. (this can also change depending on if using a so called high end water cooler)
So 440 potential watts give or take on a 504 watt PSU (I don't even figure in anything but the 12 volt rail)
87% on a power supply that is not really what I would call high quality= Not what I would even consider on a system that cost that much ( THIS WE CAN ALL AGREE ON)
Most quality power supplies can output more than their rated capacity.
The load on the PSU does not hurt the PSU only HEAT.
2 or 3 points drop on the efficiency (again nothing that concerns me)
I usually go for the 66% (2/3) range under full load unless upgrades are planed, so we will have to agree to disagree their. Again 75% I would not be stressing over since for your average user your not going to be running it that high.
For the original poster 42A= 504 watts. (on the tell tell 12 volt rail)
For the processor 95 Watts (this can vary depending on what the motherboard defaults the Vcore to)
Some of the Asus GTX 1080ti's using a torture test can peak out at 285 watts (not a realistic load but could happen)
http://www.guru3d.com/articles_pages/asus_rog_strix_geforce_gtx_1080_ti_review,7.html
For the rest of the PC I always figure 60 watts. (this can also change depending on if using a so called high end water cooler)
So 440 potential watts give or take on a 504 watt PSU (I don't even figure in anything but the 12 volt rail)
87% on a power supply that is not really what I would call high quality= Not what I would even consider on a system that cost that much ( THIS WE CAN ALL AGREE ON)
- Dec 26, 2012
- 30,756
- 4,559
- 142,790
Cpu 100w
Gpu 375w (2x8pin + pcie)
Mobo+pc 100w
Total =675w for 100% load at good OC levels. Impossible to obtain, the cpu/gpu will not reach 100% simultaneously, with running fans, SSDs, HDDs etc also at 100%. Heavy gaming loads might possibly reach 70% of max which = 473w. A 650w quality psu will see 73% at best.
73% under heavy gaming loads is chump change to a quality psu and many games won't see 70% pc usage, closer to 50% at best, so with a quality 650w psu you are looking at efficiency ranges of @50-70% when gaming.
Considering that the gpu won't use close to 100w of what the power pins say is available, closer to 275w than 375w, the figures get even better, putting you closer to 40-60% range, which is the best range for psu health, thermal properties, efficiency and output stability.
After spending over $2000 on a pc, with over half that on the gpu alone, a $60-$120 quality psu is very cheap insurance.
Will the TT work? Yes, technically it's big enough. Will it last? Doubtful. Will it deteriorate other components or kill them outright when it goes? Probably.
Gpu 375w (2x8pin + pcie)
Mobo+pc 100w
Total =675w for 100% load at good OC levels. Impossible to obtain, the cpu/gpu will not reach 100% simultaneously, with running fans, SSDs, HDDs etc also at 100%. Heavy gaming loads might possibly reach 70% of max which = 473w. A 650w quality psu will see 73% at best.
73% under heavy gaming loads is chump change to a quality psu and many games won't see 70% pc usage, closer to 50% at best, so with a quality 650w psu you are looking at efficiency ranges of @50-70% when gaming.
Considering that the gpu won't use close to 100w of what the power pins say is available, closer to 275w than 375w, the figures get even better, putting you closer to 40-60% range, which is the best range for psu health, thermal properties, efficiency and output stability.
After spending over $2000 on a pc, with over half that on the gpu alone, a $60-$120 quality psu is very cheap insurance.
Will the TT work? Yes, technically it's big enough. Will it last? Doubtful. Will it deteriorate other components or kill them outright when it goes? Probably.
Basically my plan is to wait till the summer when the 2080 comes out which will most likely be much more efficient than a 1080/1080Ti. My main use for the card in general will be playing PUBG/Fortnite maxed out anyways with no overclocking onesoever.
TRENDING THREADS
-
-
News Windows 11 Insider Build addresses mouse stuttering, Auto HDR issues, other annoyances- Started by Admin
- Replies: 4
-
Discussion PSU recommendations and power supply discussion thread - Tom's hardware- Started by Darkbreeze
- Replies: 2K
-
Discussion What's your favourite video game you've been playing?- Started by amdfangirl
- Replies: 4K
-
News AMD exec reveals Ryzen 9 9950X3D, Ryzen 9 9900X3D gaming performance — similar to Ryzen 7 9800X3D
- Started by Admin
- Replies: 28
-
Facebook
Twitter
Instagram