Thanks tool, but I just ordered a new q9650 to replace it. I'm putting that q9550 on ebay tonight. Ocing the 9550 wasn't as fun as the 9650.
This one will be on air so we'll see what I can get out of it.
Printable View
Thanks tool, but I just ordered a new q9650 to replace it. I'm putting that q9550 on ebay tonight. Ocing the 9550 wasn't as fun as the 9650.
This one will be on air so we'll see what I can get out of it.
A couple of tasters of mine so far:
Small FFTs @ 1.25vcore & 1.35vtt:
http://i194.photobucket.com/albums/z...eenShot002.jpg
Now at 56Kfft & still going :)
Highest FSB so far:
http://i194.photobucket.com/albums/z...0509x91MPI.jpg
Pardon me but why use Prime for testing? isn't OCCT Linkpack or LinX the best choice to check rock solid stability, haven't had much time testing.
http://img255.imageshack.us/img255/6954/specs2.png http://img255.imageshack.us/img255/s...ng/1/w1280.png
Some computers can pass Prime but can't pass LinX. Some can pass LinX but fail at Prime. Some pass Prime Small FFTs but fail the Blend test. Some pass the 32 bit tests but fail when running the same test when using a 64 bit OS.
And a really good computer can pass any damn test you throw at it even when grossly overclocked.
Here's the latest version of RealTemp for you.
http://www.fileden.com/files/2008/3/...alTempBeta.zip
Thanks for that Unclewebb, you're quite right. I always found I needed more vcore to pass P95 than linx on my Q6600, might be different on 45nm but I don't wanna run temps too high atm anyway.
Another good plus point about p95 is you can determine what voltages you need to increase if a certain fft size fails. Following that, I normally do a quick 1 hour test on OCCT to see if it's stable.
It can often give an error whilst doing this, but from that point it doesn't take too long to fine-tune the system, then I'll do a few hours in OCCT to finish. If I feel comfortable doing it, I'll give it a pew passes of linx too, but I'm gonna be very wary of it as I know 45nm are a little more fragile and I certainly dont want to fry or degrade the chip.
Currently at 494x8.5 to give 4.2GHz on 1.25vcore. This chip is chugging along nicely, will post pics when done :D
Still a learning curve for me, I see people mentioning VID, please tell me what does my VID mean, estimate how much voltage I'll need for the Vcore, currently its set to auto in the bios, thanks.
http://img691.imageshack.us/img691/720/vid.png http://img691.imageshack.us/img691/vid.png/1/w1280.png
1.1625 VID is good, the lower the better (mine is 1.2). Although this may not precisely determine how much vcore you'll need for a given overclock it is likely it'll be nice and low :)
If you're current setting is stable, try lowering the vcore a bit more to see if it can do the same with less.
I'm back at it with a new chip.
I replaced my E8400 with a Q9550 a few days ago.
It turns out the VID was 1.3000 on that.
I had a really hard time getting to 4000mhz.
So, I have it selling on ebay, and I ordered a new Q9650 from newegg. Turns out my luck isn't getting much better.:upset: Got it tonight.
Q9650 Batch L932B844 Packed 10/8/09... About a month ago.
VID.... 1.275v
Totally sucks.
But it seems more stable than my old one (1.25vid) was.
Here we go with the screen shots... Looks like Intel's using a spiffy new box for it's ultra crappy VID chips... LOL. tear.:cussing:
Box shots and chip:
http://i213.photobucket.com/albums/c...4/IMG_0347.jpg
http://i213.photobucket.com/albums/c...4/IMG_0345.jpg
http://i213.photobucket.com/albums/c...4/IMG_0351.jpg
IBT first. prime will follow :(
3600mhz
http://i213.photobucket.com/albums/c...sibtstable.png
3800mhz
http://i213.photobucket.com/albums/c...sibtstable.png
3900mhz
http://i213.photobucket.com/albums/c...sibtstable.png
4000mhz
http://i213.photobucket.com/albums/c...sibtstable.png
Mine arrived in the same flimsy box. Compared to the older retail chips I was unimpressed with the way it semed to rattle around in there. Mine was an L929 batch, so just 3 weeks difference. Good luck with it ;)
Temps are looking pretty good, considering this one's on air. Priming at 4200 right now.
http://i213.photobucket.com/albums/c...mestable8h.png
tought it would be nice to post it here where Q9650's fly high :D
my GTX285 vgpu/ocp/ovp mod @ 810/1836/3024 clocks i play crysis 64 on W7x64
Q9650@ 4707
bench on XP32
http://img259.imageshack.us/img259/3...0183615123.png
Cheers all :up:
Sergio
Nice Scores Sergio! Submit your results to support the team. :up:
Great job Sergio!
Thanks. Trying to do it on air this time. My sights are set a lot lower this time around.
Sergio, great results ! Very good for a single card :up:
What were ur settings for 523x9? I can't seem to get it to boot into windows. Furthest i got was near the end of loading screen. But then after trying more voltage on vcore, vtt, mch, pll, vdimm and i/o it freezes on post. I had the amps on 1000mV for CPU and MCH, and 50ps on MCH skew. :shrug:
BTW my PSU is 750w thermaltake, could that be limiting me? the 12v rail is 12.36v and the 3.3v is 3.24v on load at high clocks...
Hey mate how are you thank you for your words :up:
I can boot at this settings, don't get scared of my volts :D still a year and running all strong :up: volts shown are what my system needs
Q9650@ 4707mhz for some SPI32runs on W7x64 (wich is harder to boot than XP32)
http://img690.imageshack.us/img690/3623/captura7.jpg
Q9650@ 4707mhz for 3Dbenchs on XP32
http://img42.imageshack.us/img42/345...0183615123.png
Q9650@ 4815mhz just booted into W7x64 to run some SPI's
http://img265.imageshack.us/img265/5...51070w7x64.png
All this i made boot volts to boot and stay stable for bench are more than if you boot at low lets say 4500 and then you use ET6. Normally i use this last method, that was done for fun with Oc Nub testing togheter
Hope it helps you.
Cheers :up:
Sergio
Wow, 523fsb, that's really flying along good, that must have took a bit of tweaking, well done ;)
I'll have to have a good, long tinker after I've rebuilt my loop this weekend. Seriously doubt I'll be able to get anywhere near stable beyond 500 with this board, but the volts needed for 490x8.5 looks promising, and hopefully won't be after a lot more for ~500. If I spot a ud3p anywhere going for a good price i'll probably have to snap it up.
thanks for this andressergio.this is going to help
me out a lot.been trying to get 4707 stable and
not having much luck because im too greedy with
my volts.im now going to be a better man and give
my board whatever it needs:D
Very nice results Sergio.:up:
Put my giga rig back together last weekend and will play with these quads some more soon. Just been taking a break from it for a while.
Very nice Sergio! Better ram would have improved your scores a lot too.... :yepp:
Some bandwidth... PL8
http://www.hostthenpost.org/uploads/...61e0af7cd7.jpg
32m PL9
http://www.hostthenpost.org/uploads/...04d45da669.jpg
Cheers!
yeah for sure mate but those are not my best benches except 3DM06 for now yes...
this is my best SPI
http://img32.imageshack.us/img32/1682/spi20.jpg
this is one shoot at 4770
http://img691.imageshack.us/img691/8...1272spi321.png
cheers mate :up:
Sergio
Hi chaps,
I was wondering if anyone else has had any issues trying to stabilise at around 56-96K FFT in prime 95? I believe I've got enough vcore for the job, and have messed with GTL settings along with vMCH/VTT/vPLL for the past day. It always resets when hitting 96k and occasionally fails the 80k prior to that. Another overnight bleand test proved stable everywhere else until it hit the 96k range. At this point temps are at their highest across the board, but don't go far over 50°C which is still lots cooler than with my Q6600 before so I don't think its overheating (Q9650 hits 62-63°).
Anyone else experienced this and know how to stabilise? This is only at 466fsb too.
I'm at a loss :/
4770 / 4905 MHz ?!?! astounding stuff!
My friend sorry to say this...
here's the problem Asus Maximus II Formula P45 Rev.A3 + Q9650 you will have that sort of crap, see the disccussions around, me and lots of user went for the UD3P easy job to up a quad at 500+
on my MF2 couldnt go stable more than 475...when my mate Jor3lBR sent me the giga 500 was a minute...
I hate to say cauz i put myself in your place...but its like that
Cheers :up:
Sergio
Yeh, I was hoping the different CPU and memory would have resolved the issue but it's clear that this pos mobo simply isn't up to the task. Getting the CPU stable at 490fsb was easy as pie, but anything that stresses the nb beyond small fft range craps out just as easily. This is even with 10+ PL, lightest clock twister and lowest memory divider.
I'm hunting down a ud3p as I type, just gotta sell some parts first then I'll hopefully get my hands on one before christmas.
I know the ud3p is renouned for it's fsb capabilities, but I'm not overly concerned about breaking any fsb records, but would like 500fsb stable. Would the DQ6 or the extreme models do just as well? I kinda like the dQ6 layout more. I would also like to have some good fan control and temperature monitoring features, which I know the ud3 range lacks, does the DQ6 offer more of these? (sorry, a bit off topic here).
from what jor and napalm have said the extreme and DQ6
are very good boards with quads too.dont know about fan
control or temp monitoring.
oh i 100% agree the UD3P is the best buy.i was just
letting him know that they do OC quads good too.
and i remember some people in europe saying the
UD3P/R's are hard to find right now.
Server 2003 is the best OS ever! :cool: Sorry, I had to say that... it's just so mean when it comes to pushing hardware. Here's a little 1 hour of prime blend at 4.4GHZ...
http://www.hostthenpost.org/uploads/...088e904d3c.jpg
Cheers
I have the DQ6 and the UD3P, both are good boards. I was able to get higher fsb stable on the dq6. The only down side is that there is no LLC option on the DQ6 so your idle vcore will have to be higher. Other than that I highly recommend the DQ6. Both boards should give you a fairly easy 500fsb so cant go wrong either way IMO.
A 1m spi run I did last week for that hwbot country comp. I wish there had been some cooler weather during that stage for some better runs but it didnt happen. :up:
http://i131.photobucket.com/albums/p...pImage5x-1.jpg
That is one very effective 1M run! Very, very good in fact considering you're on DDR2. :up:
My best run with the Rampage Extreme-E8600-DDR3 was 9.344sec at 4940MHz. (my former PB- crappy E8600)
You guys are having so much fun on this P45/ 9650 combi it almost wants to make me get some. :)
here's mine stable with CPU vcore: 1.39v
http://img269.imageshack.us/img269/9642/oc1m.jpg
http://img36.imageshack.us/img36/8828/oc32m.jpg
http://img509.imageshack.us/img509/2206/oclinx.jpg
Verry nice i was priming at 1.216V load 500 x 8 with my Q9550 but it gets to 70C at all cores and i have a Z600 so i think i need some new coolpaste
Thanks. That was done on XP32 sp3 although I was planning on getting a copy of server03 to see if it will do any better. :up:
No FSB records broken here, but after installing a pair of half-decent modules, I've finally managed to get completely stable at 480x9 / 1150 PL8 with 4gb of this chaintech stuff - surprisingly good sticks of memory - only 1.86v needed of their recommended 2.1v for rated timings :).
Current Vcore of 1.4 is way overkill and I reckon I could drop at least .05v considering I could do 4.17GHz with 1.25v. I'm am yet to plot how this CPU scales with voltage, just wanted to rule out any cpu instability whilst I push the mobo & memory...
http://i194.photobucket.com/albums/z...ble1153PL8.jpg
*Edit:
Managed to get vcore down to 1.3125v. That's not bad for 4.32GHz is it?
Now I'm looking at pushing to 490 :)
Loco23,thanks for the app again.
and congrats on hwbot score1very nice.
This is my Q9650 OC took me some time to get it stable with a 2k FSB with 8GB of Ram. :up:
http://i227.photobucket.com/albums/d...B-8GBofRam.png
khaldum, which mobo did you use?
quick question.. a good 9650 how much voltage needs for 4ghz(aircooled)? linx stable?
I used the Gigabyte EP45-UD3P v1.1 Bios Version F9.
The Voltage I used is 1.325 in the Bios which seems to be lower than I've seen on most 4GHz 9650's
Using 8GB of G.Skill 1066 DDR2.
Pretty much the same here ;):
http://img21.imageshack.us/img21/9559/skrmklipp2.jpg
http://img21.imageshack.us/img21/9764/skrmklippv.jpg
Hi Papatsonis!
I've a Q9650 from the same batch as yours with same VID of 1,1250V...
Currently i'm Prime64 and Linx64 stable at 4Ghz with 1,21V :D on a DFI Lanparty X48 LT T2R under water (Swiftech Apex Ultra kit)...
Very nice chip, but warm...
Water needed! ;)
Bye
My X48 grunny:rofl: keeps walking. I wish i only had a lowv. chip to threw in it and it would easily do 555x9 - as you see i like the round numbers:rofl::rofl:
http://img99.imageshack.us/img99/4493/500x91200pl7v.jpg
:rofl::rofl::rofl:
i be back
low vid chip commin my way sooooon
andressergio please say gigabut not gigabyte:p:
I prefer 'Arseus' :D
Mine is the UK spelling :rofl:
HI, I am old and new and n00b and I come back :D:D from reading to posting
Past 6 months I watched and reading ;);) and I learned a lot from you, (especially from NapalmV5, andressergio, Jor3lBR, SNiiPE_DoGG, radaja...)
and now the time has come to post something:
I think my processor is not as good as yours but
- 1,2875 vid
- DQ6 has huge vdroop (at high FSB --> 0,10-0,12V)
- F12 bios
- boot in windows with 4,50 GHz
- anything above 520-525 FSB --> freeze or restart system
- stable at 4,00 GHz (445 FSB) for now
Here are some results:
max CPU-Z
http://valid.canardpc.com/cache/banner/807231.png
SuperPi
http://www.imagesforme.com/thumb/thumb_ad720a5d.jpg
wPrime
http://www.imagesforme.com/thumb/thumb_27291308.jpg
3Dmark06
http://www.imagesforme.com/thumb/thumb_34a88823.jpg
24/7 for now
http://www.imagesforme.com/thumb/thumb_023a5936.jpg
PS.
Sorry for bad english (google translate do miracles)
:toast:
Im starting getting bored with my grunny sexy ARSLUS... Im thinking to go for 10 teenie gigaBUTTS too
:rofl::rofl:
cheers:toast:
@greg , for 450fsb the RF how much MCH needs? (with x38-dq6 , 450fsb/1200ram8gb , needs max vmch-1.625 for questionable stability... , for 445/1187, just 5 mhz only lower.. is 9hrs linx stable with 1.525)
Damn, it sucks seeing you people with gigabyte boards pushing the hell out of these things. It's a pain to reach 4.25ghz even with my MIIF, although doable. It's not worth it to upgrade for a 250mhz increase in clocks for 24/7, although it would be very nice to bench at 4.75ghz+ Ah well, 775 is getting old. Shouldn't waste more money it I guess. I'll upgrade to I7 once I get some spare money laying around :p:
E-mil,very nice clocks and results.it takes work
to get results like your,so i know you've done the work.
jor told me the same thing once, "read up and test
like crazy to find out what your particular cpu/MB/ram
combo likes and needs".great job:up:
GUYS, how much nbv needs someone with X38-DQ6 @ 450fsb + 1200 8gb ram
Ihave a friend with maximus formula X38 and he needs crazy nbvolts to run his system stable @ much lower clocks - something like 1.69-1.73. But the thing is that he runs his system with these crazy volts for over 2 years now.
My guess is that you can safely run with up to 1.70nbv IF you can keep nb cool.
no man.. its not dreaming... right now i have finished stability tests(9hrs Linx64 - 7200mb) for 445 fsb/1.2vtt , and 8gb gskill 8800pi@ 1187 PL6 with 1.525 VMCH , just the chipset scales heavily non linear.. and for 450 fsb i need 1.625 VMCH (havent test all possible gtlref settings for 450 though)
forgot to tell , with Turbo mode, not Extreme (for extreme need ~0.1+ volt additional the x38) , extreme saves about 2.5ms in latency and give a small boost in read , but doesn't worth the heat...
neither mine (pi 8800) can go above 1220-1230 , till there are stable with minimum voltage from dq6 (1.89) but almost dont scale at all, had tried till 2.05 and saw increasing errors after 1.98v
I haven't tried to find vcore needed for 4GHz, but for 466x9=4.194GHz I need 1.25v, and for 480x9=4.32GHz, 1.3125v is needed. My vid is 1.2v but vcore needed is likely to be very similar judging by the way it scales. I'll have a stab later, just busy getting 4.4GHz stable with just 1.35v :cool:
Guys, today I removed my CPU and I noticed 2 pads on the CPU didn't have any pin marks, is that normal? I was getting some slow downs in prime95 so maybe that's related?
I marked them on this picture, they are the 14th and 15th pads from the right? what does these 2 pads are responsible for?
http://i47.tinypic.com/sgqpea.jpg
Small update :)
wPrime @ 4662 MHz (validation)
http://www.imagesforme.com/thumb/thumb_eb8328ba.jpg
^memory time sucks.
Yes, it really seems they have identical behavior! :D
I'm running it now @ 4Ghz with 4GB (2x2) of OCZ PC2 8500 Platinum Edition @ 1:1 (DDR2 891) with:
- timings 4-4-4-12,
- strap 333/667,
- Performance Level 7
with only 1.21VCore, 1.20VTT, 1.40V on MCH and 1.85V on DRAM.
Everest cache and memory benchmark average marks in these values:
- Read: 9045 MB/s
- Write: 9465 MB/s
- Copy: 8520 MB/s
- Latency: 55.0 ns
I've done accurate fine tuning on GTL+ and memory timings to gain stability with these low voltages and i think that i'll be able to handle this overclock even on summer... :up:
DFI boards are so tricky to be fine tuned but when you have time and persistence it's a satisfactory goal! :yepp:
Bye
nice!! awesome vMCH , ONLY 1.41!!!, either my X38 chipset , or the DQ6, cant do these speeds at such low voltage , or i miss some magical setting :/ , my stable are 445/1187 PL6 with 1.525 vMCH :/ , ive done extensive testing on all possible GTL adjustments (in fact only GTLRef 1 , cause the oher GTL except normal, refuses to boot)with no practical gain :( . Relaxing PL, didnt bring any substancial gain (perhaps, 0.025 vmch less - so no point) , tried to relax the x38, testing with 4gb ram only, neither that worked..
my q9650 is at 4.095Ghz - 455*9 @ 1.25v - ive got an asus p5q deluxe and 2*2gb corsair dominator 1066 ram - my cpu is 20 loops minx stable and 6hours p95 stable with cpu at 4.095 and ram at 1092mhz - all voltages have obviosly been increased...
for some reason when i got over 455 even by 1mhz it becomes unstable in games BSOD's on me an restarts my pc when running linx!!!! what u recon i need to increase to get it stable at 460mhz???
vcore - 1.256v gtl refs 0.660 + 0.660
cpu pll - 1.56v
vtt - 1.26v
vram - 2.14v
vNb - 1.24v
any help guys??
cheers
Since a lot of you are pushing your Q9650's for quite awhile now, what's the voltage you guys have been able to use consistently without getting overclock degradation? (on air).
Has anyone used something like 1.45v over long periods of time?