malevolent
can into space
Legendary
 Offline
Activity: 3472
Merit: 1721
|
 |
February 15, 2012, 12:07:02 AM |
|
Thanks for clarifying but I'm still not sure how much load can my circuit sustain, saw 16A labels on my sockets, I'll have to ask the owner of the house  btw. made a mistake, 16A not 15A |
Signature space available for rent.
|
|
|
|
|
|
Even in the event that an attacker gains more than 50% of the network's
computational power, only transactions sent by the attacker could be
reversed or double-spent. The network would not be destroyed.
|
|
|
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction.
|
|
|
|
|
DeathAndTaxes
Donator
Legendary
Offline
Activity: 1218
Merit: 1079
Gerald Davis
|
|
February 15, 2012, 12:23:10 AM |
|
Thanks for clarifying but I'm still not sure how much load can my circuit sustain, saw 16A labels on my sockets, I'll have to ask the owner of the house btw. made a mistake, 16A not 15A The CIRCUIT is 16A. Go to the circuit breaker turn off one 16A breaker. Every outlet that stops working is on THAT CIRCUIT. The Circuit can't have more than 16 amps. It doesn't care if you put 16A in one outlet or 5 amps in 3 outlets and 1 amp in a 4th. When the entire circuit >16A the breaker will trip and everything (which may be a half dozen outlets) connected to that circuit turns off. |
|
|
|
|
|
cuz0882
|
|
February 25, 2012, 02:00:12 AM |
|
Is there still no word on the release date? Hopefully the fan design looks more like the 5970, I still don't even see any pictures released yet.
|
|
|
|
|
rjk
Sr. Member
Offline
Activity: 448
Merit: 250
1ngldh
|
|
February 25, 2012, 02:20:53 AM |
|
Hopefully the fan design looks more like the 5970,
Why? The better design is the 6990, which doesn't cook the second GPU by blowing hot air over it. |
|
|
|
DeathAndTaxes
Donator
Legendary
Offline
Activity: 1218
Merit: 1079
Gerald Davis
|
|
February 25, 2012, 02:23:48 AM |
|
Hopefully the fan design looks more like the 5970,
Why? The better design is the 6990, which doesn't cook the second GPU by blowing hot air over it. That's debatable. Dumping hot air back into the intake path of other cards so same hot air can be recycled over and over and over. Would be nice to see AMD use a out the back design but a split air path where air is split half passing over each core. |
|
|
|
|
rjk
Sr. Member
Offline
Activity: 448
Merit: 250
1ngldh
|
|
February 25, 2012, 02:26:17 AM |
|
Hopefully the fan design looks more like the 5970,
Why? The better design is the 6990, which doesn't cook the second GPU by blowing hot air over it. That's debatable. Dumping hot air back into the intake path of other cards so same hot air can be recycled over and over and over. Would be nice to see AMD use a out the back design but a split air path where air is split half passing over each core. I guess I get stuck in the thinking of open frame miners, and keep forgetting that there are people that used enclosed cases  Ideally the GPUs and their heatsinks would be offset so that the air would blow over them in parallel. But that is difficult is such a small space. EDIT: I just saw that you already said that. Great minds think alike?  |
|
|
|
DeathAndTaxes
Donator
Legendary
Offline
Activity: 1218
Merit: 1079
Gerald Davis
|
|
February 25, 2012, 02:29:06 AM |
|
Hopefully the fan design looks more like the 5970,
Why? The better design is the 6990, which doesn't cook the second GPU by blowing hot air over it. That's debatable. Dumping hot air back into the intake path of other cards so same hot air can be recycled over and over and over. Would be nice to see AMD use a out the back design but a split air path where air is split half passing over each core. I guess I get stuck in the thinking of open frame miners, and keep forgetting that there are people that used enclosed cases Ideally the GPUs and their heatsinks would be offset so that the air would blow over them in parallel. But that is difficult is such a small space. EDIT: I just saw that you already said that. Great minds think alike? Yeah but I doubt they will do that. Most "sane" people buy 1 or less 7990 so the recirculation isn't that much of an issue. I think ATX standard needs to be updated. Nobody really needs 8 expansion slots but they could use 5 expansion slots spaced wider apart. |
|
|
|
|
irobindotse
Newbie
Offline
Activity: 17
Merit: 0
|
|
February 27, 2012, 11:59:58 AM |
|
I think ATX standard needs to be updated. Nobody really needs 8 expansion slots but they could use 5 expansion slots spaced wider apart. [/quote] Amen for that!  |
|
|
|
|
this time
Newbie
Offline
Activity: 55
Merit: 0
|
|
February 27, 2012, 08:08:32 PM |
|
No way a 7990 gets 1350 mh/s. We can dream though.
|
|
|
|
|
DeathAndTaxes
Donator
Legendary
Offline
Activity: 1218
Merit: 1079
Gerald Davis
|
|
February 27, 2012, 08:14:51 PM |
|
7970 can get you 675+ Mh/s and you can buy them today for $560, that is $0.83 per Mh
IF (a big IF) 7990 can be pushed to run at 1350 Mh/s, the price per hash works out to $0.63 per Mh. Even then, 5970 is a king (if you have it).
5970 does about ~1.7x a 5870. A 6990 does about ~1.7x a 6970. Thinking you will get 2.0x performance from a dual board is unrealistic. The last two dual boards (5970 & 6990) used 3 phase power instead of 4 phase power. The VRMs, power connectors and cooling systems were not designed to handle 200% of single card performance. Sure you can overclock & overvolt a 5970 but you can overclock and overvolt a 5870 higher. 675 MH/s * 1.7 = 1150 MH/s and even that is likely pushing it. I think 1 to 1.1 GH/s is more likely for stable 24/7 operation. Still the first "gigahash" class board is pretty cool in its own respect. |
|
|
|
|
|
dizzy1
|
|
February 28, 2012, 01:29:37 AM |
|
They was someone on this board claiming that they were getting 800 mh/s from a 7970 that was watercooled. If you can cool the card effectivtly I think you may be able to get to 1.3 gh/s. But it will hit at least 1.0 gh/s. Though the power used is going to be huge though.
|
|
|
|
|
|
waterboyserver
|
|
February 28, 2012, 04:05:17 AM |
|
Yeah I was able to reach 825 MH/s on a full EK waterblock at about 1300 MHz core. The voltage needs to be around +1.1V for stability, and artifacts begin to appear significantly at that speed, anything higher than 1350 MHz crashes fast on 1.1V (1375 MHz = 50% overclock, crashes in an instant, unless doing max overvolt, but not feasible for many obvious reasons) on watercooling. On air cooling artifacts appear around 1225 MHz (~70 C), watercooling around 1300 MHz (~32 C); temperature appears to be factor in this, but I won't get away with this for long since winter is ending, and no more freezing outside air will be chilling my 200mm radiator.
Without a doubt the 7990 will reach over 1 GH/s, but whether a 7990 could attain close to double the hashing rate of what I reach on my 7970 I would not be easily convinced, unless someone can accomplish it through trial (even if any modding is needed to be stable). The major issues to consider is the amount of heat a single 7970 can produce just through its own vrm; one could speculate that a 7990 would draw close to 500 watts on a single board. Thus the amount of electrical current needed is extremely high, and cooling would probably be the top concern for an overclocked hashing 7990; besides when you begin to overclock, the current also rises up to a point, until which the voltage will need to be raised. I am just wondering how a significantly overclocked 7990 would stand on air cooling: temperature-wise, power consumption-wise, and then noise-wise. I guess all we can do is wait and see how this upcoming video card performs once it is released.
|
|
|
|
|
|
cuz0882
|
|
February 28, 2012, 04:28:11 AM |
|
Hopefully they will put 3x 8 pins on the 7990. There are cards out with 3x 8 pins, I would think it could reach 7970 speeds if it was built that way. I can't get any more hashing power out of my 6970's then my 6990. They both run at 440hashs, if I raise them above 950mhz they because unstable for things like gaming. At higher clocks like the 7990 should be able to do, max power draw should be a issue with 2x 8 pins.
|
|
|
|
|
|
Mousepotato (OP)
|
|
February 28, 2012, 09:40:41 PM |
|
I heard a rumor that March 6th is the release date. Anybody know if this is true?
|
Mousepotato
|
|
|
|
bulanula
|
|
February 28, 2012, 10:13:16 PM |
|
I heard a rumor that March 6th is the release date. Anybody know if this is true?
Yeah. That is what I heard too. I still think getting 1.3 ghash/s out of a 7990 will be pretty easy. After all, getting 440 on a 5870 = easy and getting 800 mhash/s ( not quite double 440 but a 7970 gets 700 easily ) on a 5970 = easy as well so ... |
|
|
|
|
|
jake262144
|
|
February 29, 2012, 02:05:30 AM |
|
I heard a rumor that March 6th is the release date. Anybody know if this is true?
Yeah. That is what I heard too. Are you guys kidding? This is the release date of the 78xx series. |
|
|
|
|
|
cuz0882
|
|
February 29, 2012, 02:14:14 PM |
|
I heard a rumor that March 6th is the release date. Anybody know if this is true?
Yeah. That is what I heard too. I still think getting 1.3 ghash/s out of a 7990 will be pretty easy. After all, getting 440 on a 5870 = easy and getting 800 mhash/s ( not quite double 440 but a 7970 gets 700 easily ) on a 5970 = easy as well so ... It's really unclear right now, no one even knows what they look like yet. Even if the card can only gets 1.1 it will be a better mining card then the 7970. I don't see anything about march 6th online anywhere. Seems like they would have announced something if it was going to be that soon. |
|
|
|
|
DeathAndTaxes
Donator
Legendary
Offline
Activity: 1218
Merit: 1079
Gerald Davis
|
|
February 29, 2012, 03:18:41 PM |
|
I heard a rumor that March 6th is the release date. Anybody know if this is true?
Yeah. That is what I heard too. I still think getting 1.3 ghash/s out of a 7990 will be pretty easy. After all, getting 440 on a 5870 = easy and getting 800 mhash/s ( not quite double 440 but a 7970 gets 700 easily ) on a 5970 = easy as well so ... It's really unclear right now, no one even knows what they look like yet. Even if the card can only gets 1.1 it will be a better mining card then the 7970. I don't see anything about march 6th online anywhere. Seems like they would have announced something if it was going to be that soon. Exact specs haven't even been "leaked" (although companies now have timelines for their "leaks") yet. No way it is March 6th. Gamers need time to commit to $850 and specs, and early previews, websites talking about they have one but are NDA until March XX, etc. None of that has happened and usually all that occurs weeks (plural) prior to release date. |
|
|
|
|
|
jake262144
|
|
February 29, 2012, 03:52:38 PM |
|
Exact specs haven't even been "leaked" (although companies now have timelines for their "leaks") yet. No way it is March 6th. Gamers need time to commit to $850 and specs, and early previews, websites talking about they have one but are NDA until March XX, etc. None of that has happened and usually all that occurs weeks (plural) prior to release date.
Roger that. What we may expect is the release date to be announced at CeBIT. I expect that date to be a few months away, though. |
|
|
|
|
|
bulanula
|
|
March 01, 2012, 12:16:58 AM |
|
Exact specs haven't even been "leaked" (although companies now have timelines for their "leaks") yet. No way it is March 6th. Gamers need time to commit to $850 and specs, and early previews, websites talking about they have one but are NDA until March XX, etc. None of that has happened and usually all that occurs weeks (plural) prior to release date.
Roger that. What we may expect is the release date to be announced at CeBIT. I expect that date to be a few months away, though. Damn it. I was getting excited about the 6 March date  They are supposedly releasing the Pitcairn chips so that may be good as well for better $/MH ratio  |
|
|
|
|
|
