I bet they tried using the 1600mhz ram divider then.
Printable View
I bet they tried using the 1600mhz ram divider then.
Great results, this keeps getting better!
il be getting a 1055t as soon as they come to uk.
any recomendations on a cooler for this beast?
ridney what cooler are you using?
great scores btw keep up the good work
i do believe he's using the water cooling from his sig.Quote:
Originally Posted by jazkat
so, the turbo doesn't work all the time running cinebench single!?
try setting on the task manager the cinebench process to one core only!?
anyway, nice cpu :cool:
What my biggest question is:
Does it beat a 1366 i7 if you disable HT?
In games, how is the minimum FPS @ low resolutions (when you're CPU bound)?
http://www.legionhardware.com/articl...d_5970,13.html
That link shows AMD Phenom @ ~100fps where a 1156 i5/7 can do 150fps.
50% improvement at low resolutions if you're on Intel. Does the 6 core AMD help this? I'd really love to know the answer to that.
Why would adding cores improve performance in software that isn't benefitted by extra cores?Quote:
Originally Posted by anthony256
Because the CPU itself is faster and benefits from the turbo mode?Quote:
Originally Posted by MrMojoZ
It's not cores, but power/efficiency.
i will do the maximum oc on friday night (singapore time), would like to get to know the chip a little bit betterQuote:
Originally Posted by FlanK3r
got it at S$308. 1090T was listed at S$448 but was still unavailableQuote:
Originally Posted by imamage
can you give me some download links?Quote:
Originally Posted by Nintendork
will tryQuote:
Originally Posted by FlanK3r
Yep! SLS at Cybermind shop :up: the guys at hardwarezone said it was available at PC Themes but for the life of me i couldn't find that shop :shrug:Quote:
Originally Posted by Macadamia
I guess it was working all the time but on different cores, depending on how the thread was bouncing on each core. I will try to re-run cinebench with 6 cpuz for each core and take some screenshotsQuote:
Originally Posted by Spectrobozo
Your the man ridney!
Take your time, and thanks for the input...
I'll be watching this thread! :up:
Friday night? Doh!
hehe thanks dave, told you i was gonna be back in action with thuban ;-)Quote:
Originally Posted by Daveburt714
because i lost my touch :) after not seeing the "cell menu" in bios for more than half a year, i was simply lost. i need to re-learn everythingQuote:
Originally Posted by richierich
I'm so looking forward getting a 1090T along with a Gigabyte GA-890FXA-UD7, should be an epic rig to replace my Phenom II X2 550. Already these early leaked results and benchmarks tell me this is a good chip from AMD.
I'm loving the speed/voltage for single threaded applications, and LOVING the two extra cores. Coming from a C2 965BE this should be a nice increase!
POV-Ray 3.7Quote:
Originally Posted by ridney
The 1055T with a hefty NB OC should be particularly impressive in POV-Ray as the benchmark is very sensitive to memory access. This is why Nehalems get a 20%+ boost from HT in it. On a 965 @ 4Ghz, taking the NB from 2000 to 2600 and timings from 9-9-9-24 to 8-8-8-20 result in a 20%+ boost. I'd expect a 1055T @ 4.2Ghz + 3Ghz NB to perform very close (6000ish) to if not actually exceed the 980X (6400).
x264 HD Benchmark
You'll need Avisynth to go with x264 HD.
3DS Max SPEC
okay guys, here are some screenshots of voltages at the bios level. it shows that at 2.8Ghz it has 1.375v vcore and 1.15v cpu-nb
under cpu specifications
http://img85.imageshack.us/img85/6483/thuban001.jpg
under hwmonitor
http://img697.imageshack.us/img697/9186/thuban0002.jpg
i tried disabling CnQ from cell menu it but it still idles at 800mhz. I think i need to set vcore explicitly in order for it to work. And there's no option there for disabling "Turbo CORE" so perhaps setting vcore would disable it. haven't tried yet
cell menu
http://img714.imageshack.us/img714/9679/thuban0003.jpg
so now i re-did the cinebench 64bit 1 CPU run and made 6 instances of cpuz for each core and made some screenshots along the way until it finished. you can now see pretty clearly which cores are on turbo and at what load level they kick-in. (sorry for the really big pics)
http://img691.imageshack.us/img691/4...tock1cpu01.png
http://img682.imageshack.us/img682/5...tock1cpu02.png
http://img177.imageshack.us/img177/6...tock1cpu03.png
http://img63.imageshack.us/img63/867...tock1cpu04.png
http://img140.imageshack.us/img140/7...tock1cpu05.png
http://img153.imageshack.us/img153/6...tock1cpu06.png
and did the multithreaded run on cinebench too
http://img361.imageshack.us/img361/8...tockxcpu01.png
http://img219.imageshack.us/img219/9...tockxcpu02.png
http://img146.imageshack.us/img146/8...tockxcpu03.png
http://img85.imageshack.us/img85/788...tockxcpu04.png
i will try to run an overclocked version of this later, my wife needs to check cafe world :D (i need to fix her dell studio)
Woo hoo man,nice update :D .Your single thread score is identical as if you have OCed your chip to 3.3Ghz and assigned only one core in affinity settings then rerun the test ,so the Turbo CORE works flawlessly,a perfect scaling from 2.8->3.3Ghz !!! And it's out of the box feature,nice going AMD :)
This is why I have 2 laptops and I'll be putting my 965 into a secondary desktop once I get my hands on an X6 =)Quote:
Originally Posted by ridney
Great results. Can't wait to see more.
Nice, looks like AMD wasn't lieing about about how quickly it will toggle turbo when threads are shifting from core to core. Thanks for the screen shots and the effort. :up:
wooot that was a detailed view. thanks
You might wanna try TPMonitor instead of opening multiple instances of cpuz
http://www.cpuid.com/pics/tmonitor_01.png
http://www.cpuid.com/tmonitor.php
Nice work ridney.:up: Thanks for taking the time to run all those tests then posting up the results.:clap: Anyother time I would call that a bunch of work but in this case it being a new toy I figure maybe it was more like a bunch of fun.:D
thanks , nice screenshots showing the turbo core swapping .
I wonder why AMD would swap the overclocked core constantly . Could it have something to do with keeping the temperature low ?
if 1 core is overclocked for turbo core , and voltage raised only for 1 core and stressed big time ... that one core would definately get much hotter compared to the other cores .
In the way they implemented it now , all cores keep somewhat the same temperature .
I'm wondering , how would the chip choose which core to bump actually ?
I believe that's an effect of the OS/benchmark. Not the CPU. When running the single-threaded CB bench the process will hop from core to core constantly.Quote:
Originally Posted by CrimInalA
There is dedicated hardware inside Thubans that dynamically and (very fast) decides which cores should be in the boosted state,according to the load on each core.
Exactly. It is reacting to scheduling decisions made by the OS, not making scheduling decisions itself as is suggested here:Quote:
Originally Posted by informal
The simplified process I'm describing is this:Quote:
Originally Posted by CrimInalA
1) The OS scheduler observes which threads have work to do and decides where (core) to execute work for each thread.
2) CPU observes the workload on each core and decides which core to boost on.
3) Repeat.