Nice, thanks for sharing. CB10 load?Quote:
Originally Posted by iocedmyself
Chip would have an nice TDP at those settings
125W * (1.2125V/1.35V)^2 * 3.6GHz/3.0GHz => 121W :yepp:
Printable View
Nice, thanks for sharing. CB10 load?Quote:
Originally Posted by iocedmyself
Chip would have an nice TDP at those settings
125W * (1.2125V/1.35V)^2 * 3.6GHz/3.0GHz => 121W :yepp:
OCworkbench is lying straight through their heads?
9000 my ass.
http://techreport.com/r.x/phenom-x4-...ine-render.gif
Macadamia: I think they used an 32bit os at ocworkbench and an 64bit os in the diagram you posted. 9950BE should get around 9570pts at 3GHz with an 32bit os. Means deneb is 10% faster here. Or under an 64bit os deneb would score round 12750pts so ~3% below qx9650.
C2 45nm jumped a lot in Cinebench... thanks to the caches.
And yes, I'm suspecting what you said. That score was 32 while the Tech Report always benches in 64.
EDIT: Still seems low. Way too low.
http://www.techwarelabs.com/reviews/.../index_5.shtml
http://www.sharkyextreme.com/hardwar...261_3756681__5
We know clock-for-clock that these Denebs are REALLY close to the Xeons in single core rendering... so wtf.
Tech Report 2.7Ghz Opteron Cinebench x64: Single core 3156.
If we take this score, scale clock linearly to 3Ghz, times 3.9 (Phenom I speedup), we get 13676. The 32 bit scores are WAY too low.
You gotta point there.Either the NB runs at 1.8Ghz(and stays this way in retail parts) or BIOS is old.We have reports older BIOSes taking Phenom II without problem even though they don't recognize them with proper model numbers(model unknown) and/or having problems with certain memory dividers.Quote:
Originally Posted by Macadamia
The Shanghai that was tested in CB had 2.2Ghz NB clock,a good bit over 1.8Ghz.
CB10 is 99,9% unaffected by the nb speed. For comparison.
MEM 533 5-5-5-15-30 unganged
CPU 2.6GHz NB 2.0GHz => 8329
CPU 3.2GHz NB 2.0GHz => 10206
CPU 3.2GHz NB 2.4GHz => 10206
MEM 533 5-5-5-15-30 ganged
CPU 2.6GHz NB 2.0GHz => 8225
Seems more affected by CAS latency
MEM 400 4-4-4-12-24 unganged
CPU 3.2GHz NB 2.4GHz => 10241
MEM 400 5-5-5-15-26 unganged
CPU 3.2GHz NB 2.4GHz => 10216
Can be they ran in ganged mode, so unganged would improve the result around 1-2%, memory and NB subsystem really has no influence here.
Thanks justapost.
Still those Deneb numbers don't correlate well with Shanghai numbers as Macadamia pointed out,so i'm not sure what is going on.
Foxconn A79-S with OCZ Reaper 4 GB and 2x 4870...just showing some clockspeed for now, CPUz screen a bit later :p
http://s10b.directupload.net/images/081209/5z6bbu2b.jpg
Not from caches but from an improved SSE engine.Quote:
Originally Posted by Macadamia
Based on the Shanghai results, scaling from 1 to multiple threads for P2 will be similar to the 45nm C2Q and not the P1.Quote:
If we take this score, scale clock linearly to 3Ghz, times 3.9 (Phenom I speedup), we get 13676. The 32 bit scores are WAY too low.
here we go - initial testing on air :D
http://s11.directupload.net/images/081210/lag4x3bg.png
So you do have one?... looks good.. any more to come? and on what kinda of air cooler.Quote:
Originally Posted by SoF
Man that looks sooooo good. :D
Can't follow you here. 2384 scored 19357pts with eight cores at 2.7GHz and 2356 scored 15372pts at 2.3GHz with eight cores on hardware info where's the 3156 single core result comming from? Assuming 100% scaling with frequency that leads to an 7% improvement for dual socket systems.Quote:
Originally Posted by informal
How much SuperPi for that 4GHz PII? Id say 17-19sec?!
My guess is 18.3 :)Quote:
Originally Posted by w0mbat
@ SOF: Could you run a cinebench 10 CPU benchmark? I'd like to see, Fritz Chess Benchmark, and x264 Benchmark if possible. That's a nice overclock on air :up: Is that a retail chip? I know CPU-Z shot doesn't say ES, just verifying.
POST SOMETHING!!:up:Quote:
Originally Posted by SoF
Id like to know what cooler aswell along with temps if you can.
Nice voltage for 4.0 if these cpu's are able to handle 1.5 well without it being out of spec/warranty range.
I do wonder how well the NB/HT will be able to overclocked and how it does with better memory.
Well that answers whether or not it will work with my mb.Quote:
Originally Posted by SoF
Did you get a NEW bios or are you on P05?
cooler was a scythe mugen with 120mm fan - 36°C bios on 1.44V. I can't tell exact bios revision right away (i am 600km away from my toy due to a business trip) but it was the same bios I used during G|C on the old phenom ;) 1m is sub 18s with stock HTT only raised cpu speed. if you would be able to run the same clockspeed with 250HTT+ it would be below 17s I say.
OK sorry for taking so long to reply just been playing with chip =D.
So it automatically defaults to 2ghz HT speed, despite what has been reported these chips already have hypertransport 3.1 or whatever the new standard is, only there won't be any boards to support them for awhile.
CB10 64bit score at 3.0GHZ (which i now have running at 1.1vcore) is 12900 with 800mhz ram multicore render and completes in just over 1minute.
btw am using an Asus m3a78-T 790GX
If i believe you, then my Q6600 at the same speed is DOOMED ?Quote:
Originally Posted by iocedmyself
http://img243.imageshack.us/img243/8474/cb10bz5.jpg
Please note that my RAM is running much higher.... ;)
Sweet numbers :)
Keep them coming guys!! :up:
So add approximatelly 10 dgrees celcius for load?
I believe thats about how it was with good cooling on phenom from idle to laod.
http://www.xtremesystems.org/forums/...1&d=1228958064
Multicore rendertime was 1minute 1 second.