Where is Cinebench R13? I heard about it few months ago (its ready) but after is silenece time ....:(
Where is Cinebench R13? I heard about it few months ago (its ready) but after is silenece time ....:(
Atleast some of the FX-8320s seem to do "OK" too.
The best FX I have previously seen has "VIDA Delta" (leakage) value of 2.283.
This chip is the FX-8350 I posted all of the power measurements with and which does 8GHz on LN2 with just 1.848V.
Value on this FX-8320 is 1.890 (lower being better), while the average at the moment is 2.3525.
The CUs are almost completely even too.
I have not fully tested it yet, however it does atleast 4.9GHz Prime95 LFFT stable with 1.438V and passes Cinebench at 5.2GHz with 1.450V. With AMD FX watercooling, in 26 celsius ambient.
:)
omg, 5200 MHz R11.5 with 1.45V?! Wow, my best chip can 5200 with 1.5V with the same cooler.
The current (I) draw of the CPU is measured under certain operating parameters with fixed cooling capacity.
Then the value is compared against the CPU fuse delta values, this will result a value which can be used to compare different parts with the same configuration.
Comparing the base frequency VIDs between the chips does not tell you the whole thruth.
Did a quick test for maximum Prime95 frequency on the chip.
I am aware it is only 5 minutes, however the FX watercooler is choking badly under such heat load.
Once the SB-TSI temperature ("motherboard") raises to 69 celcius the weakest unit starts crashing.
I even tried swapping the intake fan of the radiator to more powerfull one, yet this made no difference what so ever.
The exhaust air is barely warm so I guess the biggest issue is the integrated pump which does not provide enough flow.
Need to source out a Phanteks PH-TC14PE, it will most likely perform better with some proper fans on it.
http://img51.imageshack.us/img51/4911/fx83205gp95.png
man, this is really crazy...5000 MHz and 1.46V...This chip must be great in benchmarks! If you will not problem with overheating, thius chip must run 5300 MHz Cinebench with 1.55V or so.
No no...
1.48375V did the trick.
http://i.minus.com/iEmy2ZQCticZy.gif
http://img830.imageshack.us/img830/1...ychosmoker.gif
Will try with VapoLS next week.
That is one hot bugger! A serious 5G contender if you can manage the heat. :jealous:
Thank you for the explanation although i have no idea what CPU fuse delta values are. :shrug:
is hot, but very good Zeus, uffff! Looks like my 3rd FX-8350 (5300 MHz R11.5), this 3rd is hot too.
I have a few questions about this.
This is at Full load correct ?
I was thinking if that is about the temperature limit or close to this with all cores loaded.
Is it about the same temperature limit with only 4 modules/CU loaded with only a single thread in each module/CU or lower since all cores are adding to the heat pumping ?
The die temperature is about -25 celsius, 6510MHz stalled before I had time to save the screenshot.
During CB the temperature is around +4 celsius and that is enough for 5.6G / 1.48V.
Should do more easily with higher voltage, however the current needs to be limited as the temperatures are not at the "zone" yet.
http://img692.imageshack.us/img692/7...20vapolscd.jpg
http://valid.canardpc.com/show_oc.php?id=2680190
Whats cooling is it? Phase?
Very likely single stage, ye. That chip looks like a keeper. Great results! :D
Holy thread resurrection :)
I am still waiting for the new memory to arrive to get cracking with LN2, so real update on the OC part of the story. In the meantime, I have done lot's of gaming with the setup running perfectly stable at 4.8GHz on 1.4V. The temperature readings from page 2 were a little off as it seems. The 17C room temp are in fact 23C, subzero temps are off too, but not by that much I guess, didnt recheck.
Zeus, voltage readings on my setup are pretty accurate, also under load. I am using 130% Overcurrent and Loadline Calibration Extreme, idle 1.392V are measured 1.397V and under load the 1.404V CPU-Z displays are 1.4V measured with 2 different multimeters just to be sure. I used the marked readout points on the board next to the 24pin plug. No idea how accurate those are, I will mearue the VRM area diretcly to see if there is any difference, but so far, no sign of the board wanting to melt down my components ^^
I got the voltage reading with DMM figured out. I guess it's not a good idea to take ground from my case.
Just did an LN2 session on Vishera yesterday, see my thread for info. Hope your chip will do better.
What rams did you order?
As a rulew of tumb, over the years I learned that taking the closest ground from what you want to measure leads to the most accurate readings.
I just saw your thread and I am kind of disappointed and a little depressed, since I had something in the 8.2GHz in my mind as a target goal, which currently is the best German score. But now, I dont think to get even close to that mark after seeing your results =/
RAMs will be Corsair Dominator Platinum DDR3-2666 CL10 (2x 4GB). Corsair offered them to me after seeing my thread - very kind :)
But let's see. If that doesnt work out, then I'd give another chip a go, depending on my budget situation. Definately cant buy an infinite amount of CPUs. I am already having trouble selling the 8320 + board.
Just because my chip sucks doesn?t mean yours will be equally as bad. Maybe yours will shine under LN2 but i?m not too sure. I think your 8300 is a high leakage chip which might go higher so don?t give up beforehand.
That?s some nice rams you got offered there, i guess they didn?t read my thread? I Actually need AMD to read my thread and send me a high leakage chip. :D
Good luck pushing things buddy.
Still are not FX-8300 avaiable in Czech republic :(...Seems, will be the same as FX-8120 95W, but I was lucky with 95W version.
yes, we have here www.alfacomp.cz , www.alza.cz (but you must preorder what you want day before), www.czc.cz and http://interlink.tsbohemia.cz/
I can recommend Alfacomp and TSbohemia
PS:and be ready for awesome girls, specially in city centrum :D (one example I uploaded)
http://i.imgur.com/G5v69K5.jpg
No updates from my side as well - sorry for the long delay. I finally want to know if my chip does any good sub sub zero, but I have been very busy with life, job, Cebit, winter coming back once more - you get the idea :(