-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
****Official GTX580 user reviews/benchmarking/overclocking thread****
- Thread starter RavenXXX2
- Start date
Associate
- Joined
- 19 Jun 2009
- Posts
- 1,306
Krowzer, I know you only just got the card but could you run a benchmark and compare it with your outgoing gtx 280?
All the reviews focus on the comparisons between more recent cards . There is not much about the older gtx 200 series which I suspect many people will be upgrading from.
All the reviews focus on the comparisons between more recent cards . There is not much about the older gtx 200 series which I suspect many people will be upgrading from.
260 - 580 now that is a worthwhile upgrade
Good idea sticking with Nvidia as it means no pain switching drivers.
You should notice the awesomeness straight away.
(I'd say the same thing to someone going from a 4850 to a 6970)
Nice disclaimer.
Better to be safe than sorry.Can anyone recommend any good benchmarks to run as the last time I ran any benchmarks was 2005 probably.
Man of Honour
The nice delivery man has just been:
Nice disclaimer.
Can anyone recommend any good benchmarks to run as the last time I ran any benchmarks was 2005 probably.
heaven bench
stalker bench
residentevil 5
lost planet 2
3Dmark vantage.
Nice! Benchmarks at stock if you please. Great if you can do some at 1680x1050.
Will try to do some later today maybe tomorrow.
Err, it is the fastest GPU on the planet, it's a fact not a fanboy comment, and yeah let's not derail too much.
how would i square up to the Sapphire Radeon HD 5970 4GB
It wins some and loses some, it does well against it considering its got two GPU's under the hood.
lol no, going to order a Zalman VF3000-F today and get them clocks up.
Getting moist raven?
lol no, going to order a Zalman VF3000-F today and get them clocks up.
Last edited:
Soldato
Ok been digging a bit, it was a long time ago I was messing with Dbs and power as it was with car hifi and I have been outta that game a while. I managed to trawl up a decent fact that its what your measuring thats the key. It was definately never going to be 10 though.
An old electronics engineer always used to say it was 6, yet a lot of others said 3, now I understand why.
Shamelessly ripped from interwebz
"I was recently giving a class on using digital audio workstations (DAWs). As I was describing how to make smooth crossfades and ramp outs, one of the students asked a question that led me into a lengthy dissertation on the nature of sound. Since this information is not widely covered, I think it is particularly useful for most picture and sound editors.
First, we need to understand our friend, the decibel. The decibel, or dB, is one tenth of a Bel. A Bel is a unit of sound named after Alexander Graham Bell. In addition to inventing the telephone, he did a great deal of research into the nature of how our ears perceive sound. As it happens, a difference of 1 dB of sound power is the smallest change our ears can perceive under ideal conditions.
Now, our ears are very sensitive. In fact, the total range of our hearing is 130 dBs, which can be expressed as a ratio of about 10 trillion to one. Because that number is so large, the dB is set to what's called a logarithmic, or log, scale to make sounds more convenient to measure. On this scale, if you double the power behind a sound you get a change of 3 dB in the power level, which is a change of 6 Db in the sound pressure level (our ears sense changes in pressure, and a change in sound pressure level, or SPL, of 6Db is a doubling).. Double the power again and it increases another 3 dB (6DB SPL), for a total increase of 6 dB (12Db SPL). As an example, let's say you have a 100-watt amp and you replace it with a 200-watt amp. Your total power increase is just 3 dB (6Db SPL). Replace that 200-watt amp with a 400-watt amp and your increase again is just 3 dB(6Db SPL). Thus, a 400-watt amp has just 6 dB more power (or 12 Db SPL) than a 100-watt amp.
The meter on a console or DAW is measuring voltage, and it takes a 6 dB increase of voltage to make the power increase by 3 dB, but that makes a 6 Db increase in sound pressure level, and a correcponding 6 Db increase on a consoleÕs VU meter. In other words, if you raise the level of a sound on a console by 6 dB, you are doubling the volume. A change of 12 dB higher would make a sound four times louder. I realize that it's a bit confusing to have these two different scales, but since we are normally working only with VU and peak meters we can just remember that a change of 6 dB is either twice or half the level of a sound."
Which makes sense to me as we were messing with car hifi amps we did need to double the power to make 3DB more power, but that was in fact 6Db SPL (what we hear).
But its certainly not 10. From what I can pick up somewhere between 6 and 10 is the perceived doubling, where as 3 and 6 are very exact measurements depending on the situation you are measuring.
so 6Db (SPL) is double
12Db (SPL) is 4x
Still the fact remains the vast majority (in fact all) sound reviews of graphics cards I have read are not conducted with any science behind them and have to be taken with a pinch of salt. The actual frequency will make it sound a lot higher or lower in volume that the true Db SPL.
By their very design PC cases are going to be poor for absorbing sound and as such a lot will pass through the case, but some also will rebound internally. But I bet one case to another could deviate quite significantly depending on how the case was filled with cables, devices etc.
Its quite amazing how some materials will absorb some frequencies, this is basically what a home cinema system is overcoming when you run a test mode with a mic. Its going to pass various frequencies via the loud speakers and measure how much can actully be "heard", and hence boost or decrease that frequency to achieve a flat tone sound.
Flat materials reflect a lot, eg a wall, a pc case side, a laminate floor, "softer" materials reflect less externally and more internally, eg a comfy sofa, a carpet. Thats why a empty uncarpetted room sounds "loud" and also why a sound studio has "egg box" type walls to reflect and absorb the sound so low reflection = more pure sound.
So take the Db readings on reviews with a pinch of salt, your individual PC and pc placement may make a significant impact on those numbers, eg if you have similar frequencies it may combine in effect to your hearing, or alternatively a card that is fairly high in real measurement may not seem so bad for you due to your PC placement. This is why people don't say the same about how noisy a card is.
An old electronics engineer always used to say it was 6, yet a lot of others said 3, now I understand why.
Shamelessly ripped from interwebz
"I was recently giving a class on using digital audio workstations (DAWs). As I was describing how to make smooth crossfades and ramp outs, one of the students asked a question that led me into a lengthy dissertation on the nature of sound. Since this information is not widely covered, I think it is particularly useful for most picture and sound editors.
First, we need to understand our friend, the decibel. The decibel, or dB, is one tenth of a Bel. A Bel is a unit of sound named after Alexander Graham Bell. In addition to inventing the telephone, he did a great deal of research into the nature of how our ears perceive sound. As it happens, a difference of 1 dB of sound power is the smallest change our ears can perceive under ideal conditions.
Now, our ears are very sensitive. In fact, the total range of our hearing is 130 dBs, which can be expressed as a ratio of about 10 trillion to one. Because that number is so large, the dB is set to what's called a logarithmic, or log, scale to make sounds more convenient to measure. On this scale, if you double the power behind a sound you get a change of 3 dB in the power level, which is a change of 6 Db in the sound pressure level (our ears sense changes in pressure, and a change in sound pressure level, or SPL, of 6Db is a doubling).. Double the power again and it increases another 3 dB (6DB SPL), for a total increase of 6 dB (12Db SPL). As an example, let's say you have a 100-watt amp and you replace it with a 200-watt amp. Your total power increase is just 3 dB (6Db SPL). Replace that 200-watt amp with a 400-watt amp and your increase again is just 3 dB(6Db SPL). Thus, a 400-watt amp has just 6 dB more power (or 12 Db SPL) than a 100-watt amp.
The meter on a console or DAW is measuring voltage, and it takes a 6 dB increase of voltage to make the power increase by 3 dB, but that makes a 6 Db increase in sound pressure level, and a correcponding 6 Db increase on a consoleÕs VU meter. In other words, if you raise the level of a sound on a console by 6 dB, you are doubling the volume. A change of 12 dB higher would make a sound four times louder. I realize that it's a bit confusing to have these two different scales, but since we are normally working only with VU and peak meters we can just remember that a change of 6 dB is either twice or half the level of a sound."
Which makes sense to me as we were messing with car hifi amps we did need to double the power to make 3DB more power, but that was in fact 6Db SPL (what we hear).
But its certainly not 10. From what I can pick up somewhere between 6 and 10 is the perceived doubling, where as 3 and 6 are very exact measurements depending on the situation you are measuring.
so 6Db (SPL) is double
12Db (SPL) is 4x
Still the fact remains the vast majority (in fact all) sound reviews of graphics cards I have read are not conducted with any science behind them and have to be taken with a pinch of salt. The actual frequency will make it sound a lot higher or lower in volume that the true Db SPL.
By their very design PC cases are going to be poor for absorbing sound and as such a lot will pass through the case, but some also will rebound internally. But I bet one case to another could deviate quite significantly depending on how the case was filled with cables, devices etc.
Its quite amazing how some materials will absorb some frequencies, this is basically what a home cinema system is overcoming when you run a test mode with a mic. Its going to pass various frequencies via the loud speakers and measure how much can actully be "heard", and hence boost or decrease that frequency to achieve a flat tone sound.
Flat materials reflect a lot, eg a wall, a pc case side, a laminate floor, "softer" materials reflect less externally and more internally, eg a comfy sofa, a carpet. Thats why a empty uncarpetted room sounds "loud" and also why a sound studio has "egg box" type walls to reflect and absorb the sound so low reflection = more pure sound.
So take the Db readings on reviews with a pinch of salt, your individual PC and pc placement may make a significant impact on those numbers, eg if you have similar frequencies it may combine in effect to your hearing, or alternatively a card that is fairly high in real measurement may not seem so bad for you due to your PC placement. This is why people don't say the same about how noisy a card is.
Last edited:
Soldato
i just folded............gtx 580 ordered
i need my pc back
From all reviews its a great card, and no matter what AMD do I cannot see any significant (read >10%) price change in the short / medium term so I wouldn't stress, I would just enjoy my new toy
Associate
Also interested in this....
Krowzer, could you also letr us know the rest of your specs.... TIA
From all reviews its a great card, and no matter what AMD do I cannot see any significant (read >10%) price change in the short / medium term so I wouldn't stress, I would just enjoy my new toy
I wouldnt be so sure the latest specs on Cayman shows it to be around 2400sp 5870 with 2gb ram and 5 extra tesselators.
If thats true, and AMD price that baby right £300-350, 580gtx will be in a world of hurt.
Some random guys post on a Chinese forum lol.