Overclock.net › Forums › Graphics Cards › NVIDIA › MSI GTX465 Twin Frozr II Golden Edition Owners Club
New Posts  All Forums:Forum Nav:

MSI GTX465 Twin Frozr II Golden Edition Owners Club - Page 391

post #3901 of 5083
Quote:
Originally Posted by Metonymy View Post

So it makes me wonder why they would sell a binned 470 as a 465 given that the hardware still has something wrong with it unless they completely isolate and bypass the broken part. And that's where my BIOS theory came in. It seems like they'd have to do something specific to bypass the sub-par part of the card so that it functions 100% as a 465 without a chance of failure.
I totally follow what you're saying, but ... again, I think you're getting locked into thinking there has to be two bad SM's ... but that really doesn't have to be the case.

I suspect that nVidia said to there partners "Oh hai, guyz, we need to get out a 465 to compete with ATI at this price point right nao", and one of the following happened:

1) MSi isolated what they thought were the 'bad shaders' on certain 470 GPU's, and permanently marked them as bad at the hardware level, and if you flash to the 465 bios it will 'see' these parts and disable them, whereas the 470 bios will not, or

2) MSi marked the 'bad shaders' as bad in the bios, in which case probably flashing back to your original bios is the only way to 'properly' take it back to a 465 (this case I personally doubt, but it's possible) or

3) MSi just took their 'worst performing' 470 boards in terms of clocks/volts, locked down any two SM's and two memory chips, and sold them as 465's.

In the last scenario (potentially) MSi found their 470's boards that (for example) wouldn't be stable at some random voltage/clock combo like, say, 750MHz at 1V, and did just that. Only they decided to do a big 'value-add' thing like slapping the limited edition copper coolers on GE cards, in order to re-coup some of their losses from selling as 465's the boards that they thought were going to be 470's ... which they were doing because nVidia asked/told them to in order for nVidia to compete w/ATI.

In which case, there ARE NOT specific 'broken shaders' or other parts, they just aren't the cream-of-the-crop performers. To use your car analogy, these are the cards that burn a quart of oil every 10K miles instead of every 50K like was the manufacturers spec.

Personally I see this as the most likely scenario of how a bunch of 470 boards came to be 465's.

Honestly if it were not this way by now I think we'd know, because there'd probably have been a whole bunch of our GE cards that malfunctioned after flashing ... but this doesn't seem to be the case.

Or, of course there's a fourth option

4) These boards are 100% fully 470's, which were simply chosen at random to become 465's because nVidia needed boards out at the 465's price point.

The thing that makes me lean against that is that we seem to need more volts pretty consistently vs. people w/'real' 470's to hit the same clocks. That's why I think that's the area in which the 'binning to become a 465' took place.

Perhaps it would help to explain things this way: A chip is, in a sense, a whole bunch of tiny, tiny traces on a piece of silicon, right? Yet something makes it so that some cards will do 800MHz on 1V, and some can't even get close, right?

Well, I would suggest to you that this is because, SOMEWHERE on the chip that can't, there is a tiny imperfection, where one trace is 35nm apart from another trace, whereas the 'spec' on the card calls for it to be 40nm away. The 'best' OC'ing cards have every single trace exactly 40nm from each other. So, it's not that the bad OC'ing card doesn't 'work', it just that it's not as 'perfectly to spec' on the surface of the chip as the one that OC's really well.

I suspect the 'flashable' 465's are simply the 'less perfect' card samples vs the real 470's, but how they 'figured this out' at the individual board level, I don't really know. But the bottom line is I don't think there really are 'bad shader clusters', and I think the existence of 465's made out of 470 boards was mostly a 'market share' kinda thing. And now that the 460's are out and are kicking butt, we'll see them go away.
Edited by brettjv - 9/20/10 at 10:28am
    
CPUMotherboardGraphicsRAM
xeon X5675 6-core @ 4.1ghz (1.29v, 20x205 +ht ) rampage iii extreme msi rx470 gaming X (the $159 budget king) 3 x 2gb corsair xms3 pc12800 (9-9-9-24-1T@1600MHz) 
Hard DriveOptical DriveCoolingOS
hynix 250gb ssd (boot), 2tb deskstar (apps),1tb... plextor px-712sa - still the best optical drive... corsair h8o v2 aio W10 home 
MonitorPowerCaseAudio
asus vw266h 25.5" (1920x1200) abs sl (enermax revolution) * single 70A rail 850w silverstone rv-03 XFi Titanium 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
xeon X5675 6-core @ 4.1ghz (1.29v, 20x205 +ht ) rampage iii extreme msi rx470 gaming X (the $159 budget king) 3 x 2gb corsair xms3 pc12800 (9-9-9-24-1T@1600MHz) 
Hard DriveOptical DriveCoolingOS
hynix 250gb ssd (boot), 2tb deskstar (apps),1tb... plextor px-712sa - still the best optical drive... corsair h8o v2 aio W10 home 
MonitorPowerCaseAudio
asus vw266h 25.5" (1920x1200) abs sl (enermax revolution) * single 70A rail 850w silverstone rv-03 XFi Titanium 
  hide details  
Reply
post #3902 of 5083
I want to experiment with replacing the stock fans, probably with 92mm. Are the stock fans speed-sensing, anyone know? Thanks!
post #3903 of 5083
Quote:
Originally Posted by rms View Post
I want to experiment with replacing the stock fans, probably with 92mm. Are the stock fans speed-sensing, anyone know? Thanks!
The fans are hooked into the boards fan header. They are 4-pin, hence they have PWM speed control (blue wire) and RPM monitoring (yellow wire). The fan speed controller is connected to the GPU's temp sensor and thus can adjust speed based on 'gpu core' temperature. The fan profile is done through the driver so it's programmable. There is no built-in temp sensor in the physical fan that adjusts speeds though, if that's what you're asking.
Edited by brettjv - 9/20/10 at 12:03pm
    
CPUMotherboardGraphicsRAM
xeon X5675 6-core @ 4.1ghz (1.29v, 20x205 +ht ) rampage iii extreme msi rx470 gaming X (the $159 budget king) 3 x 2gb corsair xms3 pc12800 (9-9-9-24-1T@1600MHz) 
Hard DriveOptical DriveCoolingOS
hynix 250gb ssd (boot), 2tb deskstar (apps),1tb... plextor px-712sa - still the best optical drive... corsair h8o v2 aio W10 home 
MonitorPowerCaseAudio
asus vw266h 25.5" (1920x1200) abs sl (enermax revolution) * single 70A rail 850w silverstone rv-03 XFi Titanium 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
xeon X5675 6-core @ 4.1ghz (1.29v, 20x205 +ht ) rampage iii extreme msi rx470 gaming X (the $159 budget king) 3 x 2gb corsair xms3 pc12800 (9-9-9-24-1T@1600MHz) 
Hard DriveOptical DriveCoolingOS
hynix 250gb ssd (boot), 2tb deskstar (apps),1tb... plextor px-712sa - still the best optical drive... corsair h8o v2 aio W10 home 
MonitorPowerCaseAudio
asus vw266h 25.5" (1920x1200) abs sl (enermax revolution) * single 70A rail 850w silverstone rv-03 XFi Titanium 
  hide details  
Reply
post #3904 of 5083
Quote:
Originally Posted by TwoCables View Post
Then why does any GTX 465's BIOS work on our cards?
well that's my point it would be written into the core bios themselves from nVidia. vendors usually just do minor tweaks to the bios for their particular set up.

Quote:
Originally Posted by brettjv View Post
In the last scenario (potentially) MSi found their 470's boards that (for example) wouldn't be stable at some random voltage/clock combo like, say, 750MHz at 1V, and did just that. Only they decided to do a big 'value-add' thing like slapping the limited edition copper coolers on GE cards, in order to re-coup some of their losses from selling as 465's the boards that they thought were going to be 470's ... which they were doing because nVidia asked/told them to in order for nVidia to compete w/ATI.

Honestly if it were not this way by now I think we'd know, because there'd probably have been a whole bunch of our GE cards that malfunctioned after flashing ... but this doesn't seem to be the case.

Or, of course there's a fourth option

4) These boards are 100% fully 470's, which were simply chosen at random to become 465's because nVidia needed boards out at the 465's price point.

The thing that makes me lean against that is that we seem to need more volts pretty consistently vs. people w/'real' 470's to hit the same clocks. That's why I think that's the area in which the 'binning to become a 465' took place.
I Agree though it might be a mixture of both. in regards to just the GE's lets just say 1500 were not up to par and the rest were chosen at random for example. It would be nice to hear from other guys who have unlocked the PNY or Point of View cards and see if they are having the same issues with volts and what not.
post #3905 of 5083
Evga has B-stock of the X58 sli for $90. I might order one just so I can run three of these in benchmarks.
House heater
(14 items)
 
  
CPUMotherboardGraphicsRAM
Xeon W5580 @ 4GHz Foxconn Bloodrage GTI Zotac GTX 980Ti ArcticStorm 12GB Corsair 2000MHz @ 1444MHz 
Hard DriveOptical DriveCoolingOS
1TB x2 Samsung F3 RAID 0 LG blu-ray burner Corsair H80 Win7 Pro 64-bit 
MonitorMonitorKeyboardPower
3x 23" LG Flatron W2361VG (sexy glossy) Hisense 55" 4k tv Logitech MX5500 Corsair 850w 
CaseMouse
CoolerMaster HAF-X Logitech revolution laser (MX5500 combo) 
  hide details  
Reply
House heater
(14 items)
 
  
CPUMotherboardGraphicsRAM
Xeon W5580 @ 4GHz Foxconn Bloodrage GTI Zotac GTX 980Ti ArcticStorm 12GB Corsair 2000MHz @ 1444MHz 
Hard DriveOptical DriveCoolingOS
1TB x2 Samsung F3 RAID 0 LG blu-ray burner Corsair H80 Win7 Pro 64-bit 
MonitorMonitorKeyboardPower
3x 23" LG Flatron W2361VG (sexy glossy) Hisense 55" 4k tv Logitech MX5500 Corsair 850w 
CaseMouse
CoolerMaster HAF-X Logitech revolution laser (MX5500 combo) 
  hide details  
Reply
post #3906 of 5083
Quote:
Originally Posted by brettjv View Post
The fans are hooked into the boards fan header. They are 4-pin, hence they have PWM speed control (blue wire) and RPM monitoring (yellow wire). The fan speed controller is connected to the GPU's temp sensor and thus can adjust speed based on 'gpu core' temperature. The fan profile is done through the driver so it's programmable. There is no built-in temp sensor in the physical fan that adjusts speeds though, if that's what you're asking.
Thx brettjv! I went ahead & blew $12 on two of these:
AVC 92mm x 25mm 4-Pin PWM Hydraulic Bearing Fan (DS09225R12HPFAF)
How to mount them(!), whether to wire in series or parallel or what, and whether it will be workable at all (probably a 4slot videocard!) who knows, but we'll see when they arrive!
post #3907 of 5083
You should have bought 80mm fans. With a little grinding, the body of the fan will fit in the recessed areas of the heatsink. To be honest, these aren't the best fans. They're small blades, thin (probably flex some), and have to spin a lot faster than a full size 80mm fan.

I'm re-testing the 1774 and will put my adjustable 120mm fan to it.

















Edited by Rangerscott - 9/20/10 at 9:37pm
House heater
(14 items)
 
  
CPUMotherboardGraphicsRAM
Xeon W5580 @ 4GHz Foxconn Bloodrage GTI Zotac GTX 980Ti ArcticStorm 12GB Corsair 2000MHz @ 1444MHz 
Hard DriveOptical DriveCoolingOS
1TB x2 Samsung F3 RAID 0 LG blu-ray burner Corsair H80 Win7 Pro 64-bit 
MonitorMonitorKeyboardPower
3x 23" LG Flatron W2361VG (sexy glossy) Hisense 55" 4k tv Logitech MX5500 Corsair 850w 
CaseMouse
CoolerMaster HAF-X Logitech revolution laser (MX5500 combo) 
  hide details  
Reply
House heater
(14 items)
 
  
CPUMotherboardGraphicsRAM
Xeon W5580 @ 4GHz Foxconn Bloodrage GTI Zotac GTX 980Ti ArcticStorm 12GB Corsair 2000MHz @ 1444MHz 
Hard DriveOptical DriveCoolingOS
1TB x2 Samsung F3 RAID 0 LG blu-ray burner Corsair H80 Win7 Pro 64-bit 
MonitorMonitorKeyboardPower
3x 23" LG Flatron W2361VG (sexy glossy) Hisense 55" 4k tv Logitech MX5500 Corsair 850w 
CaseMouse
CoolerMaster HAF-X Logitech revolution laser (MX5500 combo) 
  hide details  
Reply
post #3908 of 5083
I thought I had my nvidia nvlddmkm.sys driver crash issue under control. Looks like I don't.


Turns out the only time I have issues is when watching a blu-ray disc with my Samsung drive.

http://www.newegg.com/Product/Produc...82E16827151199

I set my graphics card clocks back to stock (for a 470). I stripped the OC off my computer. I had originally thought the 258 drivers were the problem, so I was 1 version back.

Even after upgrading to the 258 drivers, I can't watch more than 2 minutes of a blu-ray disc in the drive without getting the Nvidia driver crash BSOD.

Using PowerDVD 8 (came with Blu-Ray drive) to watch the movies.

I can watch a regular DVD in the drive just fine (while still using PowerDVD 8).

Also, I could watch blu-ray discs in this drive with my MSI Hawk 5770 without issue.

Any ideas as to what the frack is going on?

***edit**

Anyone else using one of these unlocked cards and a blu-ray drive to watch movies? Which drive do you have? Are you having any issues?

I've done all necessary testing and it has something directly to do with this video card and playing blu-ray discs in the Samsung drive I have.
Edited by Metonymy - 9/20/10 at 10:35pm
Typhoon
(13 items)
 
  
CPUMotherboardGraphicsRAM
Phenom II x6 1055t 95W Gigabyte 890FX-UD5 MSI gtx465 GE (#0304) 4x2GB Gskill ECO 1.35V 
Hard DriveOSMonitorPower
WD Black 640 AALS Win7 Ultimate x64 Samsung 40" LCD TV :) Thermaltake Toughpower XT 750W 
Case
Antec P183 
  hide details  
Reply
Typhoon
(13 items)
 
  
CPUMotherboardGraphicsRAM
Phenom II x6 1055t 95W Gigabyte 890FX-UD5 MSI gtx465 GE (#0304) 4x2GB Gskill ECO 1.35V 
Hard DriveOSMonitorPower
WD Black 640 AALS Win7 Ultimate x64 Samsung 40" LCD TV :) Thermaltake Toughpower XT 750W 
Case
Antec P183 
  hide details  
Reply
post #3909 of 5083
Quote:
Originally Posted by brettjv View Post
I totally follow what you're saying, but ... again, I think you're getting locked into thinking there has to be two bad SM's ... but that really doesn't have to be the case.

I suspect that nVidia said to there partners "Oh hai, guyz, we need to get out a 465 to compete with ATI at this price point right nao", and one of the following happened:

1) MSi isolated what they thought were the 'bad shaders' on certain 470 GPU's, and permanently marked them as bad at the hardware level, and if you flash to the 465 bios it will 'see' these parts and disable them, whereas the 470 bios will not, or

2) MSi marked the 'bad shaders' as bad in the bios, in which case probably flashing back to your original bios is the only way to 'properly' take it back to a 465 (this case I personally doubt, but it's possible) or

3) MSi just took their 'worst performing' 470 boards in terms of clocks/volts, locked down any two SM's and two memory chips, and sold them as 465's.

In the last scenario (potentially) MSi found their 470's boards that (for example) wouldn't be stable at some random voltage/clock combo like, say, 750MHz at 1V, and did just that. Only they decided to do a big 'value-add' thing like slapping the limited edition copper coolers on GE cards, in order to re-coup some of their losses from selling as 465's the boards that they thought were going to be 470's ... which they were doing because nVidia asked/told them to in order for nVidia to compete w/ATI.

In which case, there ARE NOT specific 'broken shaders' or other parts, they just aren't the cream-of-the-crop performers. To use your car analogy, these are the cards that burn a quart of oil every 10K miles instead of every 50K like was the manufacturers spec.

Personally I see this as the most likely scenario of how a bunch of 470 boards came to be 465's.

Honestly if it were not this way by now I think we'd know, because there'd probably have been a whole bunch of our GE cards that malfunctioned after flashing ... but this doesn't seem to be the case.

Or, of course there's a fourth option

4) These boards are 100% fully 470's, which were simply chosen at random to become 465's because nVidia needed boards out at the 465's price point.

The thing that makes me lean against that is that we seem to need more volts pretty consistently vs. people w/'real' 470's to hit the same clocks. That's why I think that's the area in which the 'binning to become a 465' took place.

Perhaps it would help to explain things this way: A chip is, in a sense, a whole bunch of tiny, tiny traces on a piece of silicon, right? Yet something makes it so that some cards will do 800MHz on 1V, and some can't even get close, right?

Well, I would suggest to you that this is because, SOMEWHERE on the chip that can't, there is a tiny imperfection, where one trace is 35nm apart from another trace, whereas the 'spec' on the card calls for it to be 40nm away. The 'best' OC'ing cards have every single trace exactly 40nm from each other. So, it's not that the bad OC'ing card doesn't 'work', it just that it's not as 'perfectly to spec' on the surface of the chip as the one that OC's really well.

I suspect the 'flashable' 465's are simply the 'less perfect' card samples vs the real 470's, but how they 'figured this out' at the individual board level, I don't really know. But the bottom line is I don't think there really are 'bad shader clusters', and I think the existence of 465's made out of 470 boards was mostly a 'market share' kinda thing. And now that the 460's are out and are kicking butt, we'll see them go away.
I believe it is just as simple as this:
  1. The majority of these were manufactured to be sold as GTX 470s.
  2. They went through the usual quality control check.
  3. They didn't meet the GTX 470 standards.
  4. Somebody said, "Hey, instead of melting these down and starting over, let's sell all of them as GTX 465s by sticking a GTX 465's BIOS on them."
  5. Somebody else responded, "Ok cool. But we don't have an even number of these to sell; we have more than 2,000, but we have less than 2,500".
  6. Then somebody else said, "That's no problem. We'll just stick a GTX 465's GPU on a bunch until we end up with 2,500 of these cards, but we'll use GTX 470 PCBs so that they all look the same."

Now, I suppose some might wonder then why a few of these cards only have 8 memory chips: well, I believe that these were manufactured later in order to make a total of 2,500. I mean, I believe what they did was they purposely manufactured GTX 465s on a GTX 470's PCB in order to end up with 2,500 of these cards as opposed to some odd number in between 2,000 and 2,500. They probably had something like 2,300-something "unlockable" cards, so they probably didn't want to destroy the 300-something just to bring it down to an even 2,000.

And so I think that's how we ended up with 2,500 of these where there seem to be a few hundred or so that are true GTX 465s, more or less.

Besides: since we don't know what really went down at MSI (and since we will probably never find out), is it really worth speculating?

Quote:
Originally Posted by krazzymoose View Post
well that's my point it would be written into the core bios themselves from nVidia. vendors usually just do minor tweaks to the bios for their particular set up.



I Agree though it might be a mixture of both. in regards to just the GE's lets just say 1500 were not up to par and the rest were chosen at random for example. It would be nice to hear from other guys who have unlocked the PNY or Point of View cards and see if they are having the same issues with volts and what not.
But I was making a point. My point is this: if any old GTX 465 BIOS will work on our Golden Edition cards, then that's enough proof for me that the BIOS that these come is not a custom BIOS. It's just a standard BIOS for the GTX 465 that they stuck on these cards so that they function as GTX 465s. They don't have a special customized BIOS or anything. If they did, then we wouldn't be able to just use any GTX 465 BIOS. Instead, we would be limited to the BIOS that our cards came with.

So, if I can use any GTX 465 BIOS on my card, then that proves that the BIOS my card came with is not some special, custom BIOS. If my card came with a custom BIOS, then I'd be absolutely limited to the BIOS that my card came with. I wouldn't even be able to use a BIOS that came with a different Golden Edition card.

Besides: how long do you suppose it would take to customize a BIOS for each and every one of these cards? That's over 2,000 cards!
Edited by TwoCables - 9/20/10 at 10:07pm
It's a computer!
(19 items)
 
  
CPUMotherboardGraphicsRAM
i5-2500K @ 4.5GHz (1.368-1.384V fixed voltage) ASUS P8P67 EVO B3 (UEFI ver. 1850) GTX 780 ASUS DirectCU II (1228 / 6300, 1.180V) G.SKILL Ripjaws X 8GB (2 x 4GB) 1866MHz, CL9 
Hard DriveHard DriveHard DriveOptical Drive
250 GB Samsung 840 EVO (C:\) 250 GB Samsung 840 EVO (D:\) 150 GB WD VelociRaptor Samsung SH-S243N 24x DVD Burner 
Optical DriveCoolingOSMonitor
Samsung SH-S203N 20X DVD Burner Thermaltake Frio Win 7 Home Premium x64 SP1 Retail AOC G2460PG (24" 1920 x 1080 144Hz G-SYNC) 
KeyboardPowerCaseMouse
Filco Majestouch 104-key Cherry MX Blues w/NKRO Corsair HX650 (Bronze, ordered on 12-12-2009) CM 690 Intellimouse Optical (1.1A) 1000Hz polling rate 
Mouse PadAudioAudio
Basic, but premium round X-Fi Titanium HD Klipsch ProMedia 2.1 (with 16 AWG Monster Cable... 
  hide details  
Reply
It's a computer!
(19 items)
 
  
CPUMotherboardGraphicsRAM
i5-2500K @ 4.5GHz (1.368-1.384V fixed voltage) ASUS P8P67 EVO B3 (UEFI ver. 1850) GTX 780 ASUS DirectCU II (1228 / 6300, 1.180V) G.SKILL Ripjaws X 8GB (2 x 4GB) 1866MHz, CL9 
Hard DriveHard DriveHard DriveOptical Drive
250 GB Samsung 840 EVO (C:\) 250 GB Samsung 840 EVO (D:\) 150 GB WD VelociRaptor Samsung SH-S243N 24x DVD Burner 
Optical DriveCoolingOSMonitor
Samsung SH-S203N 20X DVD Burner Thermaltake Frio Win 7 Home Premium x64 SP1 Retail AOC G2460PG (24" 1920 x 1080 144Hz G-SYNC) 
KeyboardPowerCaseMouse
Filco Majestouch 104-key Cherry MX Blues w/NKRO Corsair HX650 (Bronze, ordered on 12-12-2009) CM 690 Intellimouse Optical (1.1A) 1000Hz polling rate 
Mouse PadAudioAudio
Basic, but premium round X-Fi Titanium HD Klipsch ProMedia 2.1 (with 16 AWG Monster Cable... 
  hide details  
Reply
post #3910 of 5083
Quote:
Originally Posted by TwoCables View Post

So, if I can use any GTX 465 BIOS on my card, then that proves that the BIOS my card came with is not some special, custom BIOS. If my card came with a custom BIOS, then I'd be absolutely limited to the BIOS that my card came with. I wouldn't even be able to use a BIOS that came with a different Golden Edition card.
Even IF each bios was custom for each card (with 2 specific SM's being disabled) as long as all the SM's actually work (evidenced by it unlocking and working okay), you'd still be able to use any other custom bios ... it might disable 2 different SM's ... but that wouldn't break it because all 7 SM's onboard actually work
    
CPUMotherboardGraphicsRAM
xeon X5675 6-core @ 4.1ghz (1.29v, 20x205 +ht ) rampage iii extreme msi rx470 gaming X (the $159 budget king) 3 x 2gb corsair xms3 pc12800 (9-9-9-24-1T@1600MHz) 
Hard DriveOptical DriveCoolingOS
hynix 250gb ssd (boot), 2tb deskstar (apps),1tb... plextor px-712sa - still the best optical drive... corsair h8o v2 aio W10 home 
MonitorPowerCaseAudio
asus vw266h 25.5" (1920x1200) abs sl (enermax revolution) * single 70A rail 850w silverstone rv-03 XFi Titanium 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
xeon X5675 6-core @ 4.1ghz (1.29v, 20x205 +ht ) rampage iii extreme msi rx470 gaming X (the $159 budget king) 3 x 2gb corsair xms3 pc12800 (9-9-9-24-1T@1600MHz) 
Hard DriveOptical DriveCoolingOS
hynix 250gb ssd (boot), 2tb deskstar (apps),1tb... plextor px-712sa - still the best optical drive... corsair h8o v2 aio W10 home 
MonitorPowerCaseAudio
asus vw266h 25.5" (1920x1200) abs sl (enermax revolution) * single 70A rail 850w silverstone rv-03 XFi Titanium 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: NVIDIA
Overclock.net › Forums › Graphics Cards › NVIDIA › MSI GTX465 Twin Frozr II Golden Edition Owners Club