Well Beep, don't feel bad. I just now got 4.3ghz with my 1100t on this CHV. 4.2ghz was a breeze with 0052 bios. 4.3 took 1.55v though. Didn't score well either. Still hoping for better. I peaked at 50*C durring the Bench.
Printable View
Well Beep, don't feel bad. I just now got 4.3ghz with my 1100t on this CHV. 4.2ghz was a breeze with 0052 bios. 4.3 took 1.55v though. Didn't score well either. Still hoping for better. I peaked at 50*C durring the Bench.
Cooling is SS unit.
Hardware optimizations are basicly better memory, higer HT and NB frequency, timing optimizations.
IMC in this 1090T is insane.
Anyway, screenshot is 100% genuine i did not edit it.
Ill rerun it again later, getting wierd since betwean 2 runs there is tons difference no idea why, it varies from 7.6 to 8.2 with only upping voltages.
I ran it with 4.3ghz since im looking for 24/7 stable settings with 1.5v not just Cinebench stable, its capable 24/7 stable 4.5ghz but i need find the settings what it will run 24/7 stable and that defined by 3hours LinX and 12hours Prime95.
Well I was unable to do 4.2 on M4A89, 4.23 @ 1.57v on CHV was good.Quote:
Originally Posted by charged3800z24
My CPU is either rediculous or I am lacking magical overclocking skills...and it's not like there's much to it. Set multi, set HT ref, set voltage, run test.
Tried this again.
4.28 on CHV with 1.62v load failed Cinebench 11.5 before it was a third of the way in.
how are temps
last chance to 4400 MHz run with my 1090T....1.55V. We will soon.
51c before crash. Motherboard sensor was 63c...it reads higher than my M4A89TD did.Quote:
Originally Posted by crazydiamond
first man with FX here....:), I must push it to 8.2 in next weeks...today is day of reviews for me, srry :)
Aircooled Noctua NH-D14, FX-8120 95W, Crosshair V Formula
Attachment 121180
first man with FX here....:), I must push it to 8.2 in next weeks...today is day of reviews for me, srry :)
Aircooled Noctua NH-D14, FX-8120 95W, Crosshair V Formula
Attachment 121180
9.46pts @ 4.9GHz, RAM@1600MHz, with good old 2600k - just to compare with a brand new Bulldozer-rig at the same frequency. It would be interesting to see which frequency is required to go past that.
Attachment 121181
http://xeizo.com/blogg/wp-content/up.../cinebench.jpg
http://valid.canardpc.com/show_oc.php?id=2042107
Flanker what are you playing at? This is meant to be your benchmark :p
http://img41.imageshack.us/img41/3031/834cl7.png
:clap:Quote:
Originally Posted by FlanK3r
Geez, gonna have to make a score chart just for you if you keep this up! :rofl:
Code:Name /Post# Score Cooling CPU MHz ScreenShot
01 FlanK3r /391 9.36 LN2 X6 1100T 5418 URL
02 FlanK3r /408 8.08 AIR X8 FX-8120 4941 URL
03 FlanK3r /284 7.78 AIR X6 1090T 4360
04 FlanK3r /348 5.24 AIR X4 975 4417
05 FlanK3r /259 4.80 AIR X4 970 4071
06 FlanK3r /52 4.76 AIR X4 965 4045
07 FlanK3r /381 4.24 AIR X4 A8-3850 3538 URL
ReKcOlNu: hehe, n1 :-)
Gappo: Im warm up :D (whats your cooling, looks nice chip...)
Just a VenX :)
What the f*ck? Only 8+ points w/ 8 cores running at 4950MHz? Don't think any bios that is bound to come out any time soon will give enough of a boost to justify the purchase of a new cpu for me at this moment. :(
~K8 IPC, at least it's better than Pentium 4.Quote:
Originally Posted by Zeus
K8 is only, what? 5, 6 years back in time? No reason to complain really. :D
ZEUS: hello to FX launch day :)...But after OC it is better, than OC Thuban in performance. Winrar is totaly crazy, u can get 6000kb/s!
Well, you guys say what you want... But I've read alot about folks saying that FX was a joke (worse than Thuban), and it sucked @ Cinebench. :p:
I've had this chip for ~3 hrs, and already I've broken the 8.0 barrier on CB 11.5.
Believe me when I tell you, I worked my butt off to try and get there with a number of Thubans... ;)
So far, all I've done is tweak the CPU Multi and HTRef. I haven't even started on the IMC or Mem clocks yet.
Much less, cranking V's (modest bumps only)...
Maybe, I got a golden chip (doubt it), but I'm pretty happy with my results so far!
I can already tell this 8120 is going to be fun to tweak, and I have NO regrets about investing $220 on the new tech...
http://i30.photobucket.com/albums/c3..._FX_CB_808.png
Granted, CineBench isn't the end all, be all, of benchmarks, but in all reality my whole system seems much snappier in general.
After reading alot of the reviews I thought I'd wasted my money...
Now that I have the chip in my rig, I'm not dissapointed at all. :D
The 8120's seem to be doing great with lower voltages :)
Gappo: I'm a wuss when it comes to voltage... :rolleyes:
It takes me forever to finaly dial in a chip because I'm such a voltage/temp wimp. :p:
I usually wind up with a nice 24/7 clock in the end though (even though it may take me 3 months). :rofl:
So many people have been dogging this chip (even AMD folks).
Is it an SB Killer? NO, but I seriously doubt it'll be the death of AMD as some folks would lead us to believe!
I actually like it!
It's been fun to OC so far, and I just started playing with it... ;)
Gratz on finally breakin 8 points Dave. :up:
If anything, these BD cpu's seem to overclock very well but you and i both know if i could get my trusty old Thuban to run at such speed, BD would get it's ass handed by it.
Don't mean to spoil the fun for you Dave so i really hope you will enjoy that chip and it's likely that if Asus will bring out a bios that supports BD for my old Crosshair IV F, i will get me one of these chips, just to see how high i can overclock it. 5GHz+ on AMD would be a sweet number, especially when it's 24/7 stable. :)
nice Dave, only 1.37 V ?!!!
Yep, Thuban@5GHz scores 9+. However your chip seems very promising, much lower vCore than all chips in the reviews.
Looking forward to see your scores going higher and higher :)
what's the DC mode ? :confused:Quote:
Originally Posted by El Gappo
i'm use to seeing ganged or unganged ?
perhaps not everything is as it seems in this chip
i'd like to point out that thuban has 50% more Instruction cache.
what's the DC mode ? :confused:Quote:
Originally Posted by El Gappo
i'm use to seeing ganged or unganged ?
perhaps not everything is as it seems in this chip
i'd like to point out that thuban has 50% more Instruction cache.
not my best, but im still sorting out this fx8120. ram is killing me so far
cooling is water , going to try the chiller on sunday
8.30
http://i31.photobucket.com/albums/c3.../830cin115.jpg
cooling? Dont forget at cooling setup for result comparsion
water , sorry . forgot to list that beforeQuote:
Originally Posted by FlanK3r
btw, how good watter? I have chance do it with Corsair H100?
my chiller setup, just no chiller turned on full bore. 6 gallons of antifreeze mix pumped through a a little giant pump. maintained 20c to learn bulldozer. tomorrow i turn on chiller full bore so more to come.Quote:
Originally Posted by FlanK3r
Well... Before you turn your chiller on, I thought I'd get a jump on ya.... :D :rofl:Quote:
Originally Posted by equinox1
(gear shown in sig/water)
http://i30.photobucket.com/albums/c3..._CBTweaked.png
Dave, do you have power consumption #s @ 5.1GHz with that vcore?
Dave, omg, this is awesome, looks, you have really gold chip! Congratulation! Really so low voltage?! Unbelievable !!! Tell me more about your watter setup :D...
Looks like a fantastic chip there Dave! :clap:
Dave,
Nice chip! I'll see what I can do Monday on regular water too. Hope I have as good a luck though...
This chip looks fun, are you glad you bought it?
nice cpu! im tempted for 8120 :D
double post
[QUOTE=Daveburt714;4974015]Well... Before you turn your chiller on, I thought I'd get a jump on ya.... :D :rofl:
(gear shown in sig/water)
nice Dave, im wondering if im overvolting a bit. ill play with your settings today and see what i get.
Guys, just to be clear, I can run little short benchies at those clocks but temps come into play real quick! :yepp:
The chip never seems to stop scaling with voltage on ambient (although I'm sure it does at some point :p:).
In fact tiny bumps offer huge improvements, but temps are hard to handle...
It'll probably kick butt on EQ's chiller settup, can't wait to see it! ;)
Just take it easy on volts.
BTW: Here's what I'm actually running for 24/7, in case you missed it in the BD OC thread.
http://i30.photobucket.com/albums/c3...101511_247.png
I don't have a Kill-O-Watt, but I'd imagine it's through the roof just based on temps (hit close to 60c just on that little bench.Quote:
Originally Posted by richierich
ok , lets digest this. nb was at 2560. everything else appears accurate in cpuz.
temps are crazy at 1.6v. my chiller will keep them at about 10c during a run.
8.61. i beat my thuban but it took 400+ mhz to do it
http://i31.photobucket.com/albums/c3.../861cin115.jpg
ok , lets digest this. nb was at 2560. everything else appears accurate in cpuz.
temps are crazy at 1.6v. my chiller will keep them at about 10c during a run.
8.61. i beat my thuban but it took 400+ mhz to do it
http://i31.photobucket.com/albums/c3.../861cin115.jpg
With SS that barely keep up with load...
http://img511.imageshack.us/img511/8...nshot002hn.jpg
i think ive found my limit, my chiller wont keep up with the temps and im really pushing volts now.
8.72
http://i31.photobucket.com/albums/c3.../872cin115.jpg
Awesome runs there Equinox:up:
Your chiller is better than my SS...Can't hold the load:(
Wow! Do you guys really need that much voltage for 5.3?
Either I was getting pretty close to where scaling stopped, or I just got lucky in the chip lottery.... :p:
BTW: EQ you can DL some nice little utilities for BD including CPU-z and Hardware monitor here:
http://www.cpuid.com/news/47-cpuid_r...bulldozer.html
Shows IMC and reports temps pretty well.
5380 took 1.6v , i tried for a valid but crashed. thanks for the cpuid link Dave. now i can see my nb speed. im doing a reload of win8, win7 64 bit and xp64 bit before i try more.
temps are high , i could hold -15c on load with my thuban but this 8120 is showing 30c higher at load at +5 to +10c at 5.3.
Was hoping Zambezi would do better.
Current Top 12 (Excluding LN2 & DICE)
CPU @ 5.39Ghz, enabled C6 to tame the heat @ idle
http://img543.imageshack.us/img543/4...nshot010pz.jpg
You might as well include DI if your including cascade that can go colder than di ;)Quote:
Originally Posted by ReKcOlNu
Nicely done :DQuote:
Originally Posted by Dumo
After a recent BD death, really not to keen on over 1.5 above zero....Wont be doing it again.
those thuban up above 7+ have north bridge speeds well over 3.0ghz.
I wasn't aware of that, shows how much I know about this stuff. :shakes:Quote:
Originally Posted by chew*
OK, here's current top 15, see my sig link for all results.
this cascade can hold 1090T @ ~ -98°C @ load.Quote:
Originally Posted by chew*
i was just testing to match a Sandybridge of a friend to see score diff :p
so yeah, DI should be included.
[QUOTE=Dumo;4975415]CPU @ 5.39Ghz, enabled C6 to tame the heat @ idle
nice run Dumo, not sure i can top that. but i will try :yepp:
Aw man, the more results I see, the more I want a BD :D
why has no one bought some ice and give us a 7ghz run of BD yet?
if hes not on ReKcOlNu's chart, he dosnt count :up:Quote:
Originally Posted by Zeus
Nice improvement with the new bios Dumo. :up:
Now where's the new bios for Sabertooth... :rofl:
I assume that's still on phase?
Raja said over on the ROG forums that the 9913 BIOS was just an improvement for 3Dmark 2001 SE. Is that an understatement? Does the chip actually get higher clocks on the new BIOS?
Thanks El:up:, wow sorry for the loss:(Quote:
Originally Posted by El Gappo
I tried air for about half an hour and realized that it wasn't gonna go nowhere near 5Ghz+ full cores benched
Thanks Dave:). Yep with SS. It seems bios 9913 has a problem with oc settings, it reverted back to default in windows most of the timeQuote:
Originally Posted by Daveburt714
Has anyone done any analysis of scaling on these results? (Sorry, I've only read thru 5 pages of this thread.) It seems pretty inconsistent. E.g., X3@3900MHz: 3.02, X4@3900MHZ: 4.65. 4.65/3.02 = 1.539, 54% higher score with only 33% more cores.
On the other hand, X1@3375: 0.82, X2@3392: 1.58, 93% higher with 100% more cores is closer to what you'd normally expect.
X6@3360: 6.00 vs X2@3392 :1.58, 3.8x higher with 3x as many cores. Again, higher than expected scaling. (Obviously there's many other system factors here, memory speed etc. so numbers aren't really comparable in many cases, but I wouldn't expect memory speeds and such to make more than 10% difference.)
X4@3510: 4.17 vs X6@3500: 6.25, 50% higher score with 50% more cores - just about perfect scaling there. Kind of implies that the rest of the systems were comparably configured.
X4@4400: 5.22 vs X6@4402: 7.89, 51% higher score with 50% more cores - still about what you'd expect.
Do we have FX scores with fewer than 8 cores to compare?
is that with thuban or FX your talking about there hyc?
I was thinking FX-81xx but it would be interesting to see one or two tests for Thuban too. I'd figure Thuban results would be perfectly uniform; for FX it will obviously depend on which cores you use (same module or different module).Quote:
Originally Posted by Manicdan
im still confused,Quote:
Originally Posted by hyc
the numbers you have in your post with the odd scaling, are they from a thbuan or fx chip?
Those numbers were pulled from the Cinebench results table. They're Athlon X2s and X4s, (and yes, X6 Thubans). http://www.xtremesystems.org/forums/...d.php?t=255323Quote:
Originally Posted by Manicdan
then what your seeing is a wide variety of tweaking and some results missing L3.Quote:
Originally Posted by hyc
going from Athlon X3 to Phenom X6 is going to have better scaling from L3.
other results, like 2-4 cores, could be because of ram or NB speed differences. some guys are experts with CB11.5 and able to get nearly half a point better than i can with the same clocks as me. and i wasnt using stock ram/NB timings either.
if everything stayed the same but the number of cores, scaling would be near perfect
EDIT: thanks for making me look at the chart again, im the second WORST thuban on the list. god i remember those days of using DDR2 and a corsair H50, lol. i think its time i post a score of 7pts, lol
OK, I guess that makes sense. So really the Phenom X4 and X6 scores are most comparable.
DG Lee's thread gives a data point for FX with only 4 cores, too bad his CB11.5 slide's scores are obscured by his watermark.
Heres hyc..Quote:
Originally Posted by hyc
http://img202.imageshack.us/img202/5...shot009feq.jpg
http://img143.imageshack.us/img143/3...nshot017ip.jpg
Thanks Dumo. But just to be certain, what if you ran your 8C and 4C tests at the exact same clock speeds? Then we should see 8C being exactly 2x 4C result, right? Or 4C result will be higher than proportional, due to no sharing overhead? What about 6C?
Never look at Cinebench reported clocks, just a hint :)Quote:
Originally Posted by hyc
Dumo ran both 8C and 4C at the same 5418MHz clock shown in CPU-Z, it just happens that one of the runs he booted @4.49GHz from BIOS and then tweaked clocks to 5418MHz.
Doh.... I see, thanks for the heads up.
So, 8.85/5.25 = 1.6857: the 4 extra cores are only worth 2.7 real cores.
Did quick setting in Bios on 9911 and ran a bench, no memory tweaks yet. Just feeling out Cores and NB.
http://3800z24.info/CHVF/fx8120-8150....8_2.6_1.6.jpg
AMD FX-8120 @ 4933 MHz, 8.12
http://img836.imageshack.us/img836/3...2020110000.png
so im a little late, but heres my submission of a 1055T from one of the earliest thuban batches
this is a 'little' better than my 3.4ghz run, lol
this is on water btw
Where's the CB and CPU-z screen shots? + What type of cooling (Air, H2O, etc?)Quote:
Originally Posted by Particle
I didn't take any picture as I wasn't doing runs for this thread when I ran it. All I've got is the recorded result in CB from when I was testing some overclocking stuff. I might run it again and take a snap if you wish.
Cooling wise, that's with a Corsair H100 (water). It wasn't possible with the Core Contact Freezer I was using before. Load temps would almost hit 90C and it would crash while the H100 keeps it <= 60C. heh
Mine is H20, thanks for adding me!Quote:
Originally Posted by ReKcOlNu
I buy new cooler, A70 from Corsair (my last air cooler ever!) and I will give at it 2x Ultra Kaze 3000 RPM, think, this will be as H100 with stock fans.
Here you go:
http://pcrpg.org/pics/misc/fx8120cb.png
:clap:Quote:
Originally Posted by Particle
Update
I am having issues with NB clocking and my RAM doesn't clock well either.
FX 8120
4.9ghz cores 1.52v
2.4ghz NB 1.2v
1600mhz MEM
Water
8.04pts
http://3800z24.info/CHVF/fx8120-8150....9_2.4_1.6.jpg
http://i453.photobucket.com/albums/q...igg_2009/g.png
Tried with my 890FX board.
Man, my 8120 is the shiz......
http://3800z24.info/CHVF/uber.jpg
Wow! :eek:Quote:
Originally Posted by charged3800z24
FX6100 on the hit list...
5Ghz 6x first slow run done
http://i168.photobucket.com/albums/u...sierra/3D5.png
8150 6C test on ss
http://img10.imageshack.us/img10/652...nshot036rx.jpg
Charged: wtf?:-D.....
charged, that is one killer result. How did you do it? Or shall I ask how the FX CPU was able to give such a great result?Quote:
Originally Posted by charged3800z24
He's just joking...
u can get this result by going to your cinebench folder : CINEBENCH_11.529\cb_ranking
next, you can improvise...
And so I learn.... Thanks for explaining. Just thought that maybe someone was able to unlock a huge amount of extra performance out of this "not so great" chip. Having said that, I'm waiting for them to be available in SA to buy one with a UD7.
Guys, I didn't do any of the such. It just gave that result. I have no idea why lol.... I ran a test again and it gave me a normal result. It is obvious a glitch though...
http://3800z24.info/CHVF/uber-eh.jpg
lol...and i thought that you just did it for joke...my bad, sorry...
Charged, you trully unlocked FX CPU!
This is how they were planned to perform, but due to agreement with Intel AMD had to throttle them down a lot ;)
Charged: or maybe...you found the true power of FX chip :-).
Did you check to see if your system clock was still showing the correct time of day?
I recall running a bench on an old machine, ages ago. It had a turbo button in it, but the machine design was stupid - the real-time clock update frequency was tied to the CPU frequency. So when you used turbo to make things faster, you couldn't measure the difference using the PC's clock, because it had counted faster as well.
More commonly, I've run into systems where a chunk of assembly code will disable interrupts while running a critical section; then the system clock loses updates and the benchmark measures much faster completion because very little time passed, according to the clock.
quick test for my fx-8150
http://img.techpowerup.org/111025/cinebench.jpg