Worst G0 Q6600 ever?

[Loser777]

It needs 1.448v to do 3.5GHz, while another chip that has the same VID only needs 1.4v.
1.504v 3.6GHz.
1.72v 3.8GHz.
1.74v 3.86GHz.

You guys got any tips for selling stuff on Ebay?

[Tonucci]

What is your cooler and PSU? Ambient temps and CPU temps under load ?

You mobo w/ 4Gb ram may be hindering your OC a bit. What Kind of error do you get and with what programs ?

[Loser777]

Xigmatek S1283 isn't causing the problem. My 650W is holding up just fine. My RAM isn't the problem because 3.6GHz needs 1.504v and that's 9x400, 800MHz is the stock speed on my RAM anyway.
I know a bad chip when I see one.

[philbrown23]

wiw! that chip does suck!

[Loser777]

It's a L733B508. Except it came in one of the new Dark Blue boxes and Pack Date Feb. 2008. IT'S A :banana::banana::banana::banana:ING REFURB. :mad: :down:

[Nasgul]

nFarce chipset? And Intel sucks? :ROTF:

Dude, in what 3rd world country have you been for the past 4 years?

I'd take an abacus before you even "try" to convince me of buying an nFarce chipset.

[1_budz]

It needs 1.448v to do 3.5GHz, while another chip that has the same VID only needs 1.4v.
1.504v 3.6GHz.
1.72v 3.8GHz.
1.74v 3.86GHz.

You guys got any tips for selling stuff on Ebay?

Why not just sell it on Craigslist? That way you pay no fees. ;)

[MikeB12]

that sounds like all 3 of mine. the best one was the one I got last Oct. and it will do 390x9@1.38v
the ones I bought this year all need around 1.4v to do 380x9...

all 3 need 1.5v to do 400x9 and pass ooct overnight..

[Loser777]

Wait, I forgot, I still got occasional crashing with this Chip at 3.6GHz... 1.51v was Stable.
I'm picking up the L722A I saw earlier. That's not all that bugs me about this chip though... It's a repack. Like I posted earlier, it had HSF marks on it and the L733 didn't match the Feb. 28 2008 Pack date.

[MikeB12]

That's bunk... one of my g0's was an oem, my worst in fact... but it was new, no marks...

[Nasgul]

I don't get it.

Just because the chip takes 0.04Vc more than others, you're going to waste time by replacing it?

Going back and forth just for a mere of a fraction of Vc shouldn't be such a big deal. Now, what "IF", what if the new chip you pick up takes 1.52v to hit the 3.5ghz mark? Will you go back and get a 3rd one?

My E6600, which I got when it was first released in 2006, it's not stable at 3.60ghz with less than 1.55v, yet I saw in the E6600 thread people getting 1.45v and hitting close to 3.80ghz, and I never (to this date) bother in changing it.

Batch or no batch, what matter is that you have 3.20ghz. Although you should/must get rid of the nFarce board.

[Loser777]

I care about how far I can get with this chip, and this has been the worst chip I've seen to date. If you don't care enough to replace it then that's you.

I am having no problems with the Nforce board, and my perception of them is still pretty solid, except for the horrible vdroop. Now, this board does 490+ FSB even with this chip, and that's good enough for me.

I know people who have tossed Opterons because they didn't go past 2,75GHz on stock volts.

[philbrown23]

I gave up on my quest for the golden Q6600, they are all either dead and gone or in someones rig and staying there because people would be dumb to get rid of a Q6600 that'll do 4ghz rock solid stable.

[Skratch]

not to knock on your board but im with the others on the chipset,Iv been threw a few nv chipsets and there junk,my go quad wouldnt get past 3.4 ghz on my stiker sli and on my p5k deluxe I can get it to over 3.8 all day,the nv chipsets are horrible and I would put money on it that its not your chip.

and the fact that your running 4gigs dosnt help either,my striker wouldnt even boot past 380 fsb with 4gigs on it,total pos.wasted 350 on that board and I will never run an nv chip set again,one I used a p35 board I was hooked,the board basically maxxed out any hardware I had,it was def not my bottle neck,I was a die hard nv fan and I gave up my sli 8800gt cards to run on a p35 board.I have a spare 8800gt on my comp desk as I type this.

[Zucker2k]

Change your username, maybe you get lucky?

[jcool]

You do know it won't last a week at 1,74V ? ;)
You can have mine if you find me a nice 45nm QX, it has never even seen 1,5V because its heat output is beyond any scale, but it does 4Ghz easily and not even needing the full 1,45V (just fed it that because I didn't feel like GTL/VTT tuning yet again).

[Furiøùs]

[QUOTE=Nasgul;3136510Although you should/must get rid of the nFarce board.[/QUOTE]


Please stfu already. do you even have a 750/780/790i? shut and dude. my board is great and runs better then any X38/X48 i've ever laid eyes or hands on.

stfu already jesus christ

[zanzabar]

Please stfu already. do you even have a 750/780/790i? shut and dude. my board is great and runs better then any X38/X48 i've ever laid eyes or hands on.

stfu already jesus christ

wow i guess that u only use gigbyte boards then, ive had a 780i to use and it was trash, 6phase just isnt enough, and then the 790i has to much internal latency


u really need digital pwms, so thats dfi or foxconn to see the difference in chipset quality (even if it is just the pwms making the difference)


and 1.5x isnt bad on the 750/780i they vdroop so thats probly 1.48 on load, and u dont have gtl or fine VTT control so you wont get a low cpu core

[Glow9]

You get what you get roll of the dice. If everyone returned their chips that didn't OC them to what they'd want we'de be paying up the ass for them. Did you say you bought a refurb?

[NBF]

It's not the worst, well... at least not the only worst....

My 745A takes the same Volts as yours Loser777, and I have tested it 5 boards, all with Intel chipset.

But have in mind that those "poor clockers" stand everything, almost imortals, and they heat much less with the high volts.

So, in fact you have to give more volts but the temps are equal to good chips with less volts.

[Loser777]

Oh, I took the chip back to the store, had them get me another, L743A...
The comp wouldn't boot at first because I didn't have the cooler mounted right >_<... I was pretty nervous about the chip...
And then I when I saw VID, I jumped up in crazy joy....
1.2125v.

[mikeyakame]

My L725B G0 is happiest between 3.35 and 3.45, where it needs < 1.40v and runs extremely cool. it does have the benefit of loving low multi high fsb combos. I've been fine tuning timings on my Rampage with it, and to say the least it can manage 42gflops of calculation power at a measly 3.35ghz. It also isn't phased by heat, I've had Linpack push it up to 78-81c all cores and still stable with 1.38v. I believe mine was one of the early batches produced by Intel.

[Skratch]

Please stfu already. do you even have a 750/780/790i? shut and dude. my board is great and runs better then any X38/X48 i've ever laid eyes or hands on.

stfu already jesus christ

I have gone threw 4 different boards and know a few on here that have aswell,there is a thread with one guy trying like 15 evga boards to get one the clocked well.NV CHIPSETS SUX JUST AS MUCH AS YOUR ADDED INPUT TO THIS THREAD,why dont you try running a quad core in your pos board and see how well it runs,better yet try having it boot up with 4 sticks of ram and see how awesome that works:ROTF:.

[Vapor]

Now now guys....two wrongs don't make a right :rolleyes:

Just hit the report button :) ;)

[mikeyakame]

If he can great, if he can't oh well. He's on the other side of a monitor. Forums are forums, and keyboard warriors will always remain keyboard warriors. Rebutting their pointless input with input aimed pointlessly is a pointless exercise. To each their own, never let another persons worthless opinion have some kind of worth by showing you actually care. ;> We all here to blow up computer parts anyway, at the end of the day does it really matter who makes that particular part we blow up.

[Solarfall]

mmm your sure that you have mounted your heatsink well and its making a good contact with the cpu right ??

[mikeyakame]

My heatsink is mounted correctly yes. Difference between clocks in sync and clocks out of sync with Linpack test is 10 - 15c. Prime won't push temps up any higher than 53-54C. When Linpack is doing a calculation at 42GFlops temp spikes to 75+ on all cores momentarily, when its doing 32GFlop temps barely hit 60. I've got it tuned to around 37.5-38.5GFlop at the moment, and temps spike to 70 briefly. It has to be the most intensive testing I've ever come across. Mind you this is the 64bit optimized compile.

[pescador]

Dude, in what 3rd world country have you been for the past 4 years?

I'd take an abacus before you even "try" to convince me of buying an nFarce chipset.

I don´t think in "3rd world" (like me in Brasil) we known less about computers and overclocking than the "1st world" guys. Even because here the price of components is more than the double that you pay in "1st world". In our jobs (at companies that came from the "1st World") our remuneration is worst and we work harder than "1st world workers".

We have less sites and magazines talking about computers, but we search and help each other a lot, even we search in forums of any nationality - like Xtreme Systems. We know that the worst (refurbished too) components, CPU batches, memmory chips, video cards, motherboards e etc., are more deployed and sold here (with no specification of refurb in package like in the "1st world").

I don't think the expression "1st World" is adequate, we are in the 21 century and the world is globalized. The humanity have to evolute and help with each other.

Even so we know that Nvidia chips are not the best in overclocking and final price to consumer.

ON TOPIC:

I think its not the batch or week, because the VID is more important, and your VID is good. Maybe is the memory. Try overclocking with only one stick to kill your doubts about the CPU.

<Sorry to my English>

[Leeghoofd]

My heatsink is mounted correctly yes. Difference between clocks in sync and clocks out of sync with Linpack test is 10 - 15c. Prime won't push temps up any higher than 53-54C. When Linpack is doing a calculation at 42GFlops temp spikes to 75+ on all cores momentarily, when its doing 32GFlop temps barely hit 60. I've got it tuned to around 37.5-38.5GFlop at the moment, and temps spike to 70 briefly. It has to be the most intensive testing I've ever come across. Mind you this is the 64bit optimized compile.

so nice to push hardware into destruction mode just for stability testing... if it's prime stable for a few hours, completes 3dbenches, completes superpi32mb and can do folding 24/7... I don't see the point in tests like Linpack and co that test and for sure degrade your hardware more rapidly and that pushes it towards new heights that you will never ever reach while running normal apps...

[zanzabar]

so nice to push hardware into destruction mode just for stability testing... if it's prime stable for a few hours, completes 3dbenches, completes superpi32mb and can do folding 24/7... I don't see the point in tests like Linpack and co that test and for sure degrade your hardware more rapidly and that pushes it towards new heights that you will never ever reach while running normal apps...

there is a point to make sure that it will run for a few minuets? just to make sure its stable but this multi hour orthos thing and line or the intel one is BS IMO, unless your making high load servers there is no point

[MikeB12]

Oh, I took the chip back to the store, had them get me another, L743A...
The comp wouldn't boot at first because I didn't have the cooler mounted right >_<... I was pretty nervous about the chip...
And then I when I saw VID, I jumped up in crazy joy....
1.2125v.


:clap:

looks like ev1 missed your post.. congrats, glad it worked out....

[mikeyakame]

linpack does in 10 60 second passes stability wise what orthos/prime do in 10 hours. if the cpu can last 10 mins of linpack hammering and return values without error, cpu / dram clock alignment is very good for the fsb speed. I wouldn't run linpack for anything more than 15-20mins at a given time. It's excellent for what it does, crunch numbers parallel and give you an idea what you need to adjust next.

After a day of screwing about test after test, I managed to get my Q6600 G0 running on my Rampage at 7 x 485FSB (3.4ghz) 6:5 divider, 333mhz strap, PL 7 (as I couldnt get PL 6 to post on 333 strap for the life of me with less than 1.8v CPU PLL, and 1.6V NB so gave up) and AI twister on Stronger (what a mission that was to get stable at these clocks). DDR2 (4x1gb) is at 1166Mhz 5-5-5-16-CR2-50 with every timing which could be adjusted, pretty damn close to perfect. Everest cache thingo shows synthetics of 51.5ns mem latency, 0.9ns L1 latency, 3.4ns L2 Latency. 9787MB/s Read, 8803MB/s Write, 9543MB/s Copy. A whole day and more to get it stable as is. Argh I'm tired, still got more FSB / Ram frequency in it, only at 1.3V FSB VTT, 1.55V NB, 2.28V DDR2, 1.4V Vcore, 1.1V SB, 1.74V CPU PLL. 1.6V SB PLL.

Still need to run Micro-Scope to do a full bit for bit memory suite, at least 2 or 3 passes. Which sucks since each pass takes at least 10-12hrs. Whether we run tests which fry our hardware in 5 mins or 5 hours our hardware will still end up fried heh ;> It's all in good fun, and you gotta pay to play. I've had this G0 chip for over a year, and it hasn't burnt out yet. I think its doing quite well. I just don't run silly high clocks or voltages on it, thrashing your cpu within the limits for short periods will be less likely to shorten its lifespan than 9x multi, 3.6ghz and using > 1.4v vcore to get there. If you modify / improve performance of any thing whether it be a PC, CAR, or even a vacuum cleaner. You limited it's life span the moment you took it outside it's original specs. It's something I'm more than happy to accept. Bbqing CPU's is still much cheaper than blowing up $10K+ engines. It's definitely a cheaper hobby.

My stability testing routine is pretty basic, i run memtest86+ or windows mem diag until i know the setup wont corrupt my system, when they pass, i boot into OS then do 2-3min prime / linpack runs to test the immediate stability of settings, reboot change more and repeat. At the end I usually run a few stressful benchmark suites and play a couple of games. If it doesn't crash its pretty stable. If it crashes i'll figure out why and fix it. I've never honestly run Prime, orthos etc for more than 45-60mins. X264 encoder does same loads and makes a good alternative, and the bonus is I get around to encoding something I've been putting off for a while.

[Loser777]

My new 1.2125VID chip is doing great... 1.33v was giving me 3.6GHz but then my vdroop killed it :(. 1.192v is doing 3GHz, and I haven't tested lower yet. I'll start doing more when I get a bolt-thru for my 1283 because I'm not using pushpins to mount that again.

[Dr_at_Home]

It needs 1.448v to do 3.5GHz, while another chip that has the same VID only needs 1.4v.
1.504v 3.6GHz.
1.72v 3.8GHz.
1.74v 3.86GHz.

You guys got any tips for selling stuff on Ebay?No, I have the worst q6600

L748A, 3.2Ghz needs 1.43v :up:

[Loser777]

I stand corrected...

What is also funny is how much I botched the install of this chip... When I removed the old chip I accidentally got a drop of old AS5 in the CPU socket... ended cleaning that up with a needle and a magnifying glass. Then I didn't mount the cooler right >_<...