I see the whispers are now starting ... NVIDIA Geforce 7 series cards

Message boards : Number crunching : I see the whispers are now starting ... NVIDIA Geforce 7 series cards
Message board moderation

To post messages, you must log in.

Previous · 1 · 2

AuthorMessage
kittymanProject Donor
Volunteer tester
Avatar

Send message
Joined: 9 Jul 00
Posts: 49043
Credit: 877,641,809
RAC: 200,311
United States
Message 1369365 - Posted: 18 May 2013, 16:00:49 UTC - in response to Message 1369342.  

... Performance is said to be around 30 percent faster than a GeForce GTX 780 ...

30 percent faster than itself, that's impressive. I shall buy one.


It's appears to be a typo within the article. My reading of the tea leaves is that it should read 30 percent faster than a GeForce GTX 680 ...

The 680 and 770 should be the same card.
I read someone flashed new software into the 680 and got a 770.

Really?
I have a pair of 680s.
I can make them 30% faster with a bit of firmware magic?
A kitty keeps loneliness away.
More meowing, less hissing. I speak meow, do you?

Have made friends in this life.
Most were cats.
ID: 1369365 · Report as offensive
Tutankhamon
Volunteer tester
Avatar

Send message
Joined: 1 Nov 08
Posts: 6704
Credit: 42,227,652
RAC: 13,200
Sweden
Message 1369441 - Posted: 18 May 2013, 18:59:40 UTC - in response to Message 1369365.  
Last modified: 18 May 2013, 19:04:41 UTC

... Performance is said to be around 30 percent faster than a GeForce GTX 780 ...

30 percent faster than itself, that's impressive. I shall buy one.


It's appears to be a typo within the article. My reading of the tea leaves is that it should read 30 percent faster than a GeForce GTX 680 ...

The 680 and 770 should be the same card.
I read someone flashed new software into the 680 and got a 770.

Really?
I have a pair of 680s.
I can make them 30% faster with a bit of firmware magic?


http://www.tomshardware.com/news/Nvidia-GTX-680-GTX-770-BIOS-Flash-Hack,22561.html

http://www.reddit.com/r/hardware/comments/1e8u6i/flash_your_geforce_gtx_680_to_the_unreleased_gtx/

http://semiaccurate.com/2013/05/13/you-can-upgrade-an-nvidia-gtx680-to-a-gtx770-with-a-simple-bios-flash/
Too much hormone treated meat.
Too much Monsanto veggies.
Too old and outdated constitution.
A crazy problem, as you Yanks use to say......

There is no God, and God never existed.
ID: 1369441 · Report as offensive
Profile jason_gee
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 24 Nov 06
Posts: 7471
Credit: 91,021,232
RAC: 11,606
Australia
Message 1369459 - Posted: 18 May 2013, 20:30:28 UTC - in response to Message 1369365.  
Last modified: 18 May 2013, 20:38:13 UTC

Really?
I have a pair of 680s.
I can make them 30% faster with a bit of firmware magic?


No, here's the debunking article. You can certainly modify/flash firmware (as the noob suckered the media with a modded 680 BIOS), but this won't really make it a 770 (even though you can make it show as anything you want in the driver...)
http://www.techpowerup.com/183936/geforce-gtx-680-can-be-flashed-to-gtx-770.html

680's upper frequency limit, with a good piece of silicon & good cooling, is determined by the max voltage, which is a hardware modification to change beyond 1175mV (See Kingpin Cooling site for more info).

If a 770 uses a revised GK104 chip, assuming it's even a real card that exists now, then it'll most likely have Turbo Boost 2, then it'll probably have different power circuitry and maybe unlockable voltage as with the TITAN, IOW hardware/firmware& software all different. Turbo boost 1 & 2 are too different.

If it's identical GK104 based (unrevised/unrefreshed/same-circuitry), then it's a 680, which cannot be 30% faster than itself (there isn't enough voltage without hardware mod)

Same chip on a new card design is possible, but then it'll have more different than simply a BIOS (like cooling & power), making it a totally different card.

Jason
"Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions.
ID: 1369459 · Report as offensive
Profile arkaynProject Donor
Volunteer tester
Avatar

Send message
Joined: 14 May 99
Posts: 4183
Credit: 53,049,088
RAC: 56
United States
Message 1369462 - Posted: 18 May 2013, 20:45:25 UTC - in response to Message 1369459.  
Last modified: 18 May 2013, 20:47:06 UTC

That article was saying that the GTX780 will be 30% faster than the GTX680.

GTX 780 is based on the GK110 chip that ended up in Titan cards while GTX 770 is based on GK104 425 an improved original GK104 28nm chip.


ID: 1369462 · Report as offensive
Profile jason_gee
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 24 Nov 06
Posts: 7471
Credit: 91,021,232
RAC: 11,606
Australia
Message 1369465 - Posted: 18 May 2013, 20:52:09 UTC - in response to Message 1369462.  

That article was saying that the GTX780 will be 30% faster than the GTX680.

GTX 780 is based on the GK110 chip that ended up in Titan cards while GTX 770 is based on GK104 425 an improved original GK104 28nm chip.


That one sounds plausible, while the 680 firmware stuff, not

"Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions.
ID: 1369465 · Report as offensive
kittymanProject Donor
Volunteer tester
Avatar

Send message
Joined: 9 Jul 00
Posts: 49043
Credit: 877,641,809
RAC: 200,311
United States
Message 1369471 - Posted: 18 May 2013, 21:07:29 UTC - in response to Message 1369459.  

Yeah, I read the article. Basically says you aren't making a new card of it just by changing the bios.
These 680s are new and under warranty for 3 years, so I'm not about to muck around there.

Both are running a nice stable OC.
From the stock 1098/1163boost to 1165/1223boost OC'd.

So, I'm happy to let them be all they can be as is.
A kitty keeps loneliness away.
More meowing, less hissing. I speak meow, do you?

Have made friends in this life.
Most were cats.
ID: 1369471 · Report as offensive
AlProject Donor
Avatar

Send message
Joined: 3 Apr 99
Posts: 1387
Credit: 257,908,004
RAC: 457,506
United States
Message 1369660 - Posted: 19 May 2013, 12:15:37 UTC

One blurb I read in one of the linked articles mentioned that the 'real' new card is actually going to be the Maxwell, and that it should be released sometime next year. It also mentioned that it will have about twice the performance (not sure if that is per watt, or other measure) than the current (780?) line-up. If this is true, I think I will probably wait it out, as there isn't that big of a gain for me at this time, and my next upgrade sweep will hopefully last me for a number of years. I can't afford to try to keep up with the tick-tock upgrade cycle that the big boys (Intel/Nvidia) are running at. 3-4 years is more my speed, make a decent investment and run 'em into the proverbial ground... ;-)

ID: 1369660 · Report as offensive
Profile trader
Volunteer tester

Send message
Joined: 25 Jun 00
Posts: 126
Credit: 4,968,173
RAC: 0
United States
Message 1370493 - Posted: 22 May 2013, 7:04:44 UTC - in response to Message 1369459.  



680's upper frequency limit, with a good piece of silicon & good cooling, is determined by the max voltage, which is a hardware modification to change beyond 1175mV (See Kingpin Cooling site for more info).

If a 770 uses a revised GK104 chip, assuming it's even a real card that exists now, then it'll most likely have Turbo Boost 2, then it'll probably have different power circuitry and maybe unlockable voltage as with the TITAN, IOW hardware/firmware& software all different. Turbo boost 1 & 2 are too different.

If it's identical GK104 based (unrevised/unrefreshed/same-circuitry), then it's a 680, which cannot be 30% faster than itself (there isn't enough voltage without hardware mod)

Same chip on a new card design is possible, but then it'll have more different than simply a BIOS (like cooling & power), making it a totally different card.

Jason

Curious my msi gtx680 lightning says i have the ability to ramp up the voltage quite high, as well as modify several other features wonder if maybe its is just a voltage issue and the bios mod is real if it would work for me
ID: 1370493 · Report as offensive
Profile jason_gee
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 24 Nov 06
Posts: 7471
Credit: 91,021,232
RAC: 11,606
Australia
Message 1370511 - Posted: 22 May 2013, 7:33:10 UTC - in response to Message 1370493.  

Curious my msi gtx680 lightning says i have the ability to ramp up the voltage quite high, as well as modify several other features wonder if maybe its is just a voltage issue and the bios mod is real if it would work for me


A 'good' 680 is generally reliably overclockable much higher than the presented stock 770 spec (whether real or fictitious) using a software tool like eVGA Precision X or MSI afterburner. Given the Lightning might have custom power circuitry, it may or may not allow the voltage to push past the standard max of 1175mV for a 680. That will depend on hardware & software limits, and if possible should not require a modified BIOS.

With the caveats about overvolting shortening the lifespan of your card in mind, I would suggest seeing what you can get from it using the loose approach described on my profile, with software tools. A 'good' reference 680 can stay under 70 degrees Celcius at full load, 1175mV and ~1200MHz+. Once satisfied the sweet spot is found, you can then explore options for using a 680 Bios editor if desired, but IMO honestly the flexibility of the software control improved so much as to render firmware mod-flashing an unnecessary risk.

Jason
"Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions.
ID: 1370511 · Report as offensive
Previous · 1 · 2

Message boards : Number crunching : I see the whispers are now starting ... NVIDIA Geforce 7 series cards


 
©2017 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.