Printable View
You'll probably get better performance with 183bclk and 1833mhz memory cyrptik.Quote:
Originally Posted by CryptiK
22x cpu multiplier, 183bclk, 5x mem multiplier, 20x uclk multiplier will give you almost exactly the same speeds, but you'll have a uclk 2x your memory speed.
I'm also kind of curious why you're sitting at 4ghz with a 990x? ;) You can probably do 4.4 without breaking a sweat!
Not sure about the high vs low BCLK thing, lower BCLK the system has less inherent latency than with higher BCLK's. Since these silly boards dont sit evenly on a given BCLK say I used 182 BCLK (~182.4 in reality) i'd be on ~4012MHz core, DDR3-1820MHz and 3640MHz uncore. Slightly higher CPU, 48 MHz less on the ram and 36 more on the uncore. I might get a little better performance. Thing is I hate uneven CPU/ram/uncore speeds and I try and make numbers that seem 'nice' and 'balanced'. That's part of the reason why its at 4005MHz. This CPU has no problem doing 4.5GHz @ 1.336v (see submission in 4.5GHz LinX club) but I dont need that much power, I probably wouldnt even notice. When I work it hard, renders/decoding/encoding/simulations etc, it takes so long anyway its a walk away come back in a few hours scenario and if it finished 15mins sooner I wouldnt be waiting for it so I see no need to push it and create more wear and tear on all the components. I also dont like running EIST etc so its full speed 24/7 and I'd rather not push it like that. With the 6 cores and 12GB ram it does everything I want it to at the current settings.
You can't turn off EIST, as shown in your CPU-Tweaker shot. ASUS mislabels EIST in BIOS and it's really the turbo disable. With R3E, it's impossible to stop your cpu from dropping the multiplier to save power.Quote:
Originally Posted by CryptiK
The point to going to 183bclk is you wouldn't be bottlenecking your memory. As it stands, sure your ram runs at 1866, but your UCLK is bottlenecking it to 1800mhz worth of performance. So as far as internal 'balance' of your cpu... 183bclk makes a lot more sense. If you jumped to the 23 multiplier and 183bclk(183.5 on my board) you get an even 4220MHz... What makes 4005 any more even/balanced than 4220? ;)
Also, you're an overclocker! We don't do it because we'll notice!:yepp::D
Yeah! :up: We do it because we can! :DQuote:
Originally Posted by zoson
BTW, there is a temperature drop expected here, so I'm looking forward to some action. :)
I'll just have to post in a different thread. :rolleyes:
Maybe we should make a high performance club thread? Like the 4.5ghz club but for low speed 100gflops, highest ht gflops, and highest non ht gflops? break it up by architecture, so there would be three categories for each cpu type?Quote:
Originally Posted by donmarkoni
The uncore portion provides it's own performance gains from higher MHz (uncore = everything else except the core) but there is certainly a performance gain to be had going from DDR3-1600MHz to DDR3-1868MHz at the same uncore of 3604MHz. It won't 'bottleneck the ram to DDR3-1800MHz' as you put it, but there is an interaction between the IMC portion of the analogue controller and the ram speed.
Also, my R3E does not drop to low multi at no load. I can watch the C-state and its in 100% speed mode all the time with a lowest calculated multiplier of 29.997x when 30x is set in BIOS.
I'm keen for an efficiency test, to do it right and to standardize for any CPU speed we'd need to calculate the GFLOPS/MHz (we can of course also have brackets of results as you suggested). What could be fun is doing something like the low clock SPI32M thread, where there is categories of say 100GLFOPS, 90GFLOPS, 80FLOPS etc and the lowest speed required to hit as close to but not over that mark, ie: the 70GFLOPS bracket would have an ideal target of 79.99 GFLOPS etc.
In absolute terms, for my 4GHz run I just did, I get 78.3962/4005.1 = 0.01957 GFLOPS/MHz, or to make it more meaningful, 19.57 GFLOPS/GHz.
Looking at the data from the last 2 pages, here's the absolute efficiency rankings:
- mr too short - 92.1319/4610.5 = 0.01998 = 19.98 GFLOPS/GHz
- donmarkoni - 91.9528/4610.8 = 0.01994 = 19.94 GFLOPS/GHz
- zoson - 91.5017/4657.9 = 0.019644 = 19.64 GFLOPS/GHz
- CryptiK - 78.3962/4005.1 = 0.01957 = 19.57 GFLOPS/GHz
- Spoolindsm - 86.6040/4504.9 = 0.019224 = 19.22 GFLOPS/GHz
The IMC is part of the uncore. The IMC bandwidth is directly related to uncore speed.
When Nehalem came out, Intel mandated 2x uncore to memory speed because unless uncore is 2x, there is not enough bandwidth to the IMC to support the full speed of your memory.
Intel relaxed this requirement to 1.5x on gulftown because hardware limitations made it impossible to reach high enough uncore speeds to accommodate the fastest memory out there. To fully utilize your memory without bottlenecking it, you must have 2x uncore. When you go to say 2.1x uncore, the performance gains drop off very quickly when memory is a primary component.
So essentially, if your uncore is less than 2x, you are bottlenecked. Of course you would see a performance gain from 1600mhz to 1866 with a 3.6ghz uncore. It's because you have extra bandwidth above 1600mhz.
That's why i get better performance at 2000mhz memory with 4ghz uncore and 4.6ghz than i do with 2176mhz memory, 3.95ghz uncore and 4.665ghz.
I think my R3E killed my 980X :upset: :dammit: :censored: :cussing:
WTF man. What happened?
Zoson - yeah I guess we are using the term bottlenecked slightly differently but I agree with you. I guess in my case though I'm about at the limit of what I'm comfortable running 24/7 at 1.250 Vtt (load), so 3800MHz uncore isn't going to be an acceptable setting for me as it will require more voltage.
Dude I'm so PO'd you have no idea. Months and thousands spent on finding this batch and this POS chip just DIES.
You know what's even funnier? I never pushed Vcore past 1.35v. QPI never went above 1.35v also. Except when I had QPI LLC disabled and QPI set to 1.38125v. I turned the system off and switched the jumper to enable QPI LLC. Turned it back on so I guess QPI got set to 1.38v actual. I set it down to 1.35v and restart, and the system keeps restarting after POST. Set Bios to defaults and on restart, Vcore was being set to 1.14375v manually by the Bios??
I'm thinking it was a combination of dieing motherboard that took the cpu with it. :banana::banana::banana::banana: man why always me...
Tried it in another board? Maybe the boards just acting funny - I hope the CPU's ok mate. If you reset BIOS it could be setting 1.14375v as a lower power state.
If it died at that voltage without some kind of spike then I'd be RMAing that at the least you can sell whatever they give you back if its not a nice batch.
OK this is a bit odd. Thought my setup was at the max for the voltages it was running, but after thinking about what zoson said I thought I'll just do a quick test. Now I played around with different BCLK and multi combos, different ram etc, but thought instead of re-establishing what everything needs for different speeds lets just crank up the uncore to 3733MHz (2 x ram) and leave everything else where it was. Only thing I did different was re-install my corsair ram fan since it was getting hot...........
Result. Appears fully stable at 3733 MHz uncore @ ~1.257 Vtt running 12GB (6x2GB) 1866C7's at DDR3-1868MHz 7-8-7-20 1T..........I wasn't expecting that, as I left Vtt where it for 3600MHz uncore and the same ram settings. I obviously had a ram temp issue affecting things so I'm going to re-investigate ram timings/vdimm and Vtt requirements now the ram is properly cooled. Bandwidth now has increased a few hundred MB/s and latency is a little lower as expected.
Everest at new settings:
http://i720.photobucket.com/albums/w...7-201T164v.png
LinX 5 run @ 9100MB at new settings (sorry guys I'm the new king of efficiency :D :P) :
http://i720.photobucket.com/albums/w...8gflopsmax.png
Highest GFLOPS = 80.3288
80.3288/4005 = 0.02005 GFLOPS/MHz = 20.06 GFLOPS/GHz
Nice clocks there Cryptik :up:. Interesting to see the role of Uncore in efficiency. Congrats though now you've made yourself the target lol
So we're going with only 5 runs of LinX to determine efficiency? You guys know right how sometimes if the OC is unstable you'll get HIGHER GFlops in LinX than you would when it's actually stable?
I love the affects of thermal dynamics though. So basically you're ram was just getting too hot? I'm sure better fans on the AirFlow module would definitely help. Reminds me of back when I was on air cooling and I'd wait 20-30 minutes before each run to let the mem cool off ha. That was the only reason I went with the EK Dominator block. Although it's restrictive for the system's overall flow rate, it does always keep the DIMMs cool. I took mine off because I didn't want to restrict flow. But now I'm thinking maybe I should put it back in.
About the 980X though, God knows what happened. I tried it in the OC as well, no go. I'm pretty sure that thing is dead from a voltage spike or something similar. But like you said at least I can RMA it and get my money back. Mann I wanted to get in on this GFlops competition :rolleyes:. Regardless of what batch I get though, I'm gonna overclock the life out of it :D
Yeah, I think it should be similar rules to the 4.5ghz club... 20 loops... and we need to standardize ram size because higher problem size = higher gflops.
Cryptik, try running with 5120, and see what kind of performance you get. Not everyone is going to have 12gb mem unfortunately.
Yeah I was thinking about how many loops we'd need to standardize it to, but since this isn't really a stability club (is it?) less should be sufficient so long as it's consistent. You can see in my run it's got about 0.0357 GFLOPS variance between the highest and lowest loops. Also 20 loops with 12GB of ram will take me until next year to complete, perhaps that's your grand plan! :p: I dunno, say 10 loops? That should be more than enough to establish efficiency and basic stability, although I must say I've always got lower GFLOPS when slightly unstable.
I will try running with less as a test, but I shouldn't be excluded for running 12GB :( We can always have installed ram as a delimiter and have 6/12GB categories.
Sorry to hear about the 980X mate, hope you get a good one in return. Intel are very prompt in their turn around time for a claim and this is one of those situations where it cannot be determined exactly what happened but it certainly wasn't abuse on your behalf.
Edit - I actually only run the fans on the airflow module at 60% duty cycle they're way too loud at full speed, and the ram actually doesnt need much airflow to stay cool. Its just warm to the touch now under full load and the fans are basically silent.
I've seen both higher and lower GFlops when not stable... If you're stable at 9216, all you gotta do is set ram size to 5120 and do 20 loops. It'll take less than an hour!Quote:
Originally Posted by CryptiK
Yeah Spool... That's some bad luck... :( Sorry for your loss.
I've run all my chips with qpi llc on and 1.35v set in bios for 24/7. It usually sits at 1.33v-1.34v in hwmon. The 980x I'm running now was apparently from Intel rma. It's not as good as the one you had, but still very nice. Hopefully you'll get one at least as good.
Re-ran it with 5000 MB/s selected. 20 passes. Max GFLOPS = 79.6098, so that gives me 79.6098/4005 = 19.88 GFLOPS/GHz. Not bad, good enough for 3rd behind Don.
http://i720.photobucket.com/albums/w...8gflopsmax.png
Zoson you got a good point about standardizing the test size. Does your guys' OS have enough free memory to test 5120MB? I'm a weirdo and like to test 'whole' gigabytes and I believe 5120MB is exactly 5GB. And how about 10 loops instead of 20? Maybe 10 loops for 12GB runs since it takes so much longer and 20 loops for 6GB?
And I'm with Cryptik on having separate categories for 6GB and 12GB. It's the fair way to do it. And I somehow accumulated 30GB of Hypers so I'm down to do some 12GB testing myself ha.
Thanks for the kind words and wishful thinking fellas. What's bothering me the most is the fact I never pushed the chip outside it's limits, and it still died. Feels like a waste. I really really hope they send me a good batch. But judging from my past experience, I'm not trying to get my hopes up. I feel Intel purposely sends THE worst batch they have on file when you do an RMA.
P.S. Cryptik - I just saw your edit about the AirFlow Module. So at 60% it keeps 12GB warm to the touch? And that's while benching right? If so it seems like you got good airflow in your case, so no need to watercool them.
And for your run I saw you had memory at 7-8-7-20. Have you tried 7-7-7-21? What's the difference in GFLOPS?
The issue with 12GB memory is you have to set 9216 or less to get full GFlops with HT on. So we can never really tell what kind of performance we could really extract out of 12GB, unless HT is disabled. Also... People with 12GB memory can easily just select 5120 and run loops... We want to get an easy to relate performance measurement against each other... Not a bunch of different categories that make it difficult to relate each other's performance!Quote:
Originally Posted by spoolindsm127
I think it's also important to set a high speed requirement, to keep everyone in the same scaling area. Since performance scales worse at higher clockspeeds. Sorry Cryptik. :( It's also so I won't have to maintain a TON of entries... 4GHz is super easy to hit nowdays.
I've kind of envisioned this club to be everything the 4.5ghz club wasn't...
This is what I had in mind:
The High Performance Club
Welcome to the club for overclockers that maintain high performance along with high frequencies!
How it works:
HT ON and HT OFF results are ranked by Peak GFlops.
Low clock 100 GFlop results are ranked as a function of GFlops/GHz(Rating).
Only the highest performance entry per category will be kept per user.
General Rules that apply to all categories:
1. You need LinX to submit an entry.
2. You must be using the latest Linpack libraries.
3. You must have at least 6GB memory installed in the system.
4. All memory channels must be in use.
5a. Bloomfield/Gulftown/Lynnfield must be above 4500MHz and 85GFlops.
5b. Sandy Bridge/Ivy Bridge must be above 5000MHz and 110GFlops.
6. CPUz must be visible. If your voltage does not show, list your vBIOS in your post.
7. CPU-Tweaker must be open showing sub timings.
Additional HT ON Rules:
1. Linx must be running/calculating the 21st or higher loop.
2a. Bloomfield/Gulftown memory size must be set to 5120, regardless of your overall memory capacity.
2b. Lynnfield/Sandy Bridge/Ivy Bridge memory size must be set to 7168, regardless of your overall memory capacity.
Additional HT OFF Rules:
1. Memory size must be set to max capacity - 1024. (So if you have 12GB, 11264MB)
2a. Linx must be running/calculating the 21st or higher loop for less than 8192MB memory.
2b. Linx must be running/calculating the 11th or higher loop for more than 8192MB memory.
Low clock 100GFlop rules:
1. HT OFF only.
2. Results must be less than 101GFlops (100.9999 is fine).
3. No more than half of your loops may drop below 100GFlops.
4. Linx must be running/calculating the 21st or higher loop.
5a. Bloomfield/Gulftown memory size must be set to 5120, regardless of your overall memory capacity.
5b. Lynnfield/Sandy Bridge/Ivy Bridge memory size must be set to 7168, regardless of your overall memory capacity.
Bloomfield/Gulftown/Lynnfield
Bloomfield/Gulftown/Lynnfield - HT ON
Rank User Name CPU Type Rating Peak GFlops Speed(MHz) 1 MrTOOSHORT i7 990x 19.98 92.1319 4610.5 2 donmarkoni i7 980x 19.94 91.9528 4610.8 3 zoson i7 980x 19.64 91.5017 4657.9 4 spoolindsm127 i7 980x 19.22 86.6040 4504.9
Bloomfield/Gulftown/Lynnfield - HT OFF
Rank User Name CPU Type Rating Peak GFlops Speed(MHz) 1 zoson i7 970 21.74 100.2570 4611.0 2 donmarkoni i7 980x 21.98 100.2397 4560.0 3 name CPU Type Rating GFlops Speed 4 name CPU Type Rating GFlops Speed
Bloomfield/Gulftown/Lynnfield - Low clock 100 GFlop
Rank User Name CPU Type Rating Peak GFlops Speed(MHz) 1 donmarkoni i7 980x 21.98 100.2397 4560.0 2 zoson i7 970 21.74 100.2570 4611.0 3 name CPU Type Rating GFlops Speed 4 name CPU Type Rating GFlops Speed
Sandy Bridge/Ivy Bridge
Sandy Bridge/Ivy Bridge - HT ON
Rank User Name CPU Type Rating Peak GFlops Speed(MHz) 1 name CPU Type Rating GFlops Speed 2 name CPU Type Rating GFlops Speed 3 name CPU Type Rating GFlops Speed 4 name CPU Type Rating GFlops Speed
Sandy Bridge/Ivy Bridge - HT OFF
Rank User Name CPU Type Rating Peak GFlops Speed(MHz) 1 name CPU Type Rating GFlops Speed 2 name CPU Type Rating GFlops Speed 3 name CPU Type Rating GFlops Speed 4 name CPU Type Rating GFlops Speed
Sandy Bridge/Ivy Bridge - Low clock 100 GFlop
You want me to burn my chip don't you :p:
Format looks good, quite hardcore, but good. I would still lilke a 12GB category, its not that uncommon to have 12GB or more ram these days and I'd like to be able to take advantage of the improved efficiency it offers. It's an efficiency club afterall and with the proposed rules, the most efficient setups cannot enter!
The HT OFF category is for everyone with >9GB memory. Since you can do full memory testing with HT OFF. :) I think it makes the most sense to allow large memory sizes only when the cpu can fully stress it correctly. People with large memory sizes should DOMINATE this category.
It's a club for high performance with high frequencies! It's to separate the men from the boys. There are so many fakes on the 4.5ghz club...
In order to correctly rank, the playing field has to be level for everyone involved. That means there must be a set speed limit and a set memory size, or we will have irregularities that make the results no longer apples to apples.
It's also not feasible to lower the frequency requirement down to 4ghz... If I did... every noob would submit an entry!!! It's so easy now days to run 4ghz that it's no longer a 'high frequency'. Entrance into this club should be VERY difficult. At a minimum, more difficult to get into than the 4.5ghz club.
I think a big issue is that I would like AMD to be able to compete in this club. But I don't know crap about AMD performance. Can current AMD cpu's even hit 100GFlops with LinX?
Should be 2 HT on categories for 6 and 12GB its only 2 groups not like its 10 or something so not much harder to administrate. Should make it also full mem capacity -1024MB.
spoolin - yeah now running ram at DDR3-1868MHz | 7-7-7-20 1T, tRFC 60 and slightly tighter subs @ 1.594v. Still have some fine tuning to do but its a bit better than previous settings, which I was limited to due to running with no fan over the ram.
It makes no sense to have a 12GB ht on category at all, since you can't ever stress your full memory amount, it is never known if your memory overclock is fully stable. The way bank interleaving works... You'll never fully engage the ddr channels. It's a flawed test, so it will not be added. This is the main problem. It doesn't actually show or prove anything.Quote:
Originally Posted by CryptiK
If intel ever fixes the bug in linpack that makes the test not actually stress your machine if you use over 9216mb ram, i'll re-evaluate this position at that time.
It's also the difference between having to maintain 8 groups, instead of 6... which is a 33% increase in work. That is significant. Especially since I want to have AMD included as well... Which will make at least another four categories once BD is out.
Nice. Your CMG kit is definitely very strong. Did you see any latency improvements in AIDA64? Also wondering what the difference is between 7-7-7-20 and 7-7-7-21.Quote:
Originally Posted by CryptiK
Zoson - I think what Cryptik is trying to say is more like two categories of <6GB and >6GB. Although I never experienced it myself, I know the bug you are referring to in LinX.
Nice. Your CMG kit is definitely very strong. Did you see any latency improvements in AIDA64? Also wondering what the difference is between 7-7-7-20 and 7-7-7-21.Quote:
Originally Posted by CryptiK
Zoson - I think what Cryptik is trying to say is more like two categories of <6GB and >6GB. I know about the bug you are referring to in LinX. All 12GB cannot be tested, but it is still more than 6GB.
Yes, I mean 9126MB with HT On should be a category. I keep saying 12GB due to that being the min amount of physical ram installed in an X58 board required to run that test. I should have been more clear.
I'm actually running my 1866C7GT kits, but they seem ok, not sure what constitutes strong with Hypers, I just think if GTX2's which were untouchable. My 2000C8GT are also pretty good, but I can't remember what they need for these speeds. 7-7-7-20 is what Hypers are comfortable running up to and above DDR3-2000MHz, you don't need to use tRAS of 21 as with DDR2 and some DDR3 as they are fine with the tighter step. Corsair rate them at 1866/7-8-7-20 so the 20 is stock timing and actually 2 steps tighter than the old rule book would state is required.
The idea is that the club would provide apples-to-apples comparisons of peoples tweaks, tricks, and hardware... Just like we can easily show that having donmarkoni's setup at the same speeds is still faster than my setup.Quote:
Originally Posted by spoolindsm127
You can't compare a 6gb run to a 9gb run.
This is why with HT off, I am enforcing all mem -1gb. So that people with 12gb can fully stress their machines and provide performance comparisons for each other.
People with 12gb can do the 5120 test and still have their performance directly apples-to-apples with the 6gb people.
The facts:
I already have a category for 12gb users, and it provides a real stress test that will allow apples-to-apples comparisons between users with more than 9216mb memory.
There's no reason why someone with 12gb ram can't just select 5120mb in linx and participate apples-to-apples in the HT on category.
Testing more than 9216 memory with HT on is bugged, and doesn't actually stress your machine correctly, so there's no reason to record these results.
Prime example:
I can run my 18gb kit at 1866mhz with HT on and 9216mb memory size. it will Linx FOREVER.
If I turn HT off, and run with the EXACT same settings but max mem, my memory overclock is unstable because now all three of my memory channels are fully engaged(literally, same exact settings, just turned HT off). Since testing of >9216mb with HT on can be used to cheat and make an overclock look stable, when it is actually not... It's not acceptable to have a category for it. I'm not saying Cryptik is cheating... But others will for sure.
I don't want this club to turn into the 4.5ghz club with a bunch of lame/fake results.
The idea is that the club would provide apples-to-apples comparisons of peoples tweaks, tricks, and hardware... Just like we can easily show that having donmarkoni's setup at the same speeds is still faster than my setup.Quote:
Originally Posted by spoolindsm127
You can't compare a 6gb run to a 9gb run.
This is why with HT off, I am enforcing all mem -1gb. So that people with 12gb can fully stress their machines and provide performance comparisons for each other.
People with 12gb can do the 5120 test and still have their performance directly apples-to-apples with the 6gb people.
The facts:
I already have a category for 12gb users, and it provides a real stress test that will allow apples-to-apples comparisons between users with more than 9216mb memory.
There's no reason why someone with 12gb ram can't just select 5120mb in linx and participate apples-to-apples in the HT on category.
Testing more than 9216 memory with HT on is bugged, and doesn't actually stress your machine correctly, so there's no reason to record these results.
Prime example:
I can run my 18gb kit at 1866mhz with HT on and 9216mb memory size. it will Linx FOREVER.
If I turn HT off, and run with the EXACT same settings but max mem, my memory overclock is unstable because now all three of my memory channels are fully engaged(literally, same exact settings, just turned HT off). Since testing of >9216mb with HT on can be used to cheat and make an overclock look stable, when it is actually not... It's not acceptable to have a category for it. I'm not saying Cryptik is cheating... But others will for sure.
I don't want this club to turn into the 4.5ghz club with a bunch of lame/fake results.
Yea my definition of strong is the memory being able to run tighter timings / lower frequencies than it's rated for without a problem. See I have tested multiple Hyper kits, couple of which were Corsair. In one kit each stick was capable of 2230MHz 8-8-8-24-1T 1.65v (real). However, they wouldn't run 7-7-7-21-1T at all. Another set, which I have run most of the time, can do min 2134MHz 8-8-8-24-1T and 2020MHz 7-7-7-21-1T.Quote:
Originally Posted by CryptiK
I thought you had a 2000MHz kit that was doing 1866MHz 7-7-7-20-1T. Which in my books is considered good. But even being a 1866MHz kit, the fact it can tighten tRCD to 7 is still very good.
Just got my replacement R3E from Asus. Same revision, 1.04G. Has a Lotes socket unlike my old board and the I/O jacks are opposite brand of what the old board were; meaning the sound jack was lotes on the old board and everything else like ps/2 and usb was foxconn. New board only sound is foxconn, everything else is Lotes. Some of the small capacitors/resistors look different in color also.
If this is a better, revised board I might just go back to it.
Hmm... I see that much has happened here since I was away...
High Performance Club, great idea! I like it. :up:
You guys did great, especially you Cryptik, :clap: but now I want my crown back! :p:
Math is a universal language, so I'll just say this: 84.3369/4.20004=20.08 (got you by two ;) )
http://img267.imageshack.us/img267/3...lub2008end.jpg
EDIT: Maybe Hi Perf Club deserves a thread of it's own? If you guys agree, we should ask Admins/Mods to migrate appropriate post since zoson's post#5505 when the idea was born, and of course make his post#5519 the first post of The High Performace Club?
BTW, I hope I won't be banned for posting X58A-OC result in R3E thread.
I've been weighing what's been said in the thread... I've made a few changes to the rules since the original post... Once I had a bit more feedback, I was going to start a new thread.Quote:
Originally Posted by donmarkoni
I was also considering looking through the 4.5ghz club and getting some results from there to 'seed' the new club with members... If there's a qualifying run there, have it apply here... etc...
I've also been wondering how AMD systems perform and if we could have them compete in this also? All the results I've seen can't even compete though... I'd like for AMD to be able to submit results... But I have no baseline for comparison...
Hi everyone. I had rampage III extreme for over a year now and never had problems but yesterday one came up. I unplugged all of the cables from the pc because I wanted to move my pc to the other side of the room and after I plugged everything back and started my pc I don't have internet connection since then. When I go to the device manager it says that Intel; 82567V-2 This device cannot start (Code 10). I read that this may be driver issue so reinstalled drivers and I even got the latest version but it still doesn't work. My others computers have the internet connection so it has to be the Intel network adapter not working. Any of you had this problem before or know what to do to fix it?
Got my 980X back from RMA a couple weeks back. The Intel Gods must have been watching over me on this one. Sent in Batch 3048A259, received Batch 3048A255. I had come across this Batch previously, and it had a lower VID than the A259. It's a Golden Batch :D.
4.5GHz 155.2x29 1862MHz 7-7-7-20-1T Uncore 3724MHz
Load (real)
VCore 1.316v-1.317v
QPI 1.293v
DRAM 1.662v
http://img221.imageshack.us/img221/1...ghz1860mhd.jpg
89.2273 GFlops / 4.50063 GHz = 19.83
P.S. Hope I also don't get banned for posting in this thread. I don't know if this is worth any consolation but I will be switching back to R3E for comparison shortly.
Good result! All results close to 20GFLOPS/GHz mean that rig is highly optimized. Above 20 is very hard. I'll try above 20 with 4.6GHz when the temps get lower.Quote:
Originally Posted by spoolindsm127
Have you tried pushing UnCore/RAM to ~4/2GHz?
It would be great to see another comparison between OC & R3E! :up:
intel rma comes through with the golden batch! LOL nice :D
I will start the HPC thread soon so we have an appropriate thread to post in/compare results.
Thanks Don :). Without optimizing the memory timings I was hardly pulling 86GFlops. And YESS, above 20 is VERY hard. I tried to surpass your result but it was requiring too much QPI so I pussied out. And above 20 as in above 20.07, so 20.08+. Very difficult to do.Quote:
Originally Posted by donmarkoni
Yes I was first going for 4/2GHz for Uncore/Ram. But after seeing how sensitive Gulfs are when pushing QPI, I was never able to stabilize 2000MHz/4000MHz. I am strictly not pushing QPI past 1.33v.
Also I believe the 'safe' uncore multi formula [ ((mem multi x 2) -2) within -1 or +2 ] needs to be followed or degradation happens to the IMC. Maybe that is why my last 980X died. I was pushing memory up to 2200MHz and keeping uncore at lowest multi. I also did a 4.5GHz LinX run with mem at 12x and uncore 26x.
Oh I am definitely doing a R3E vs OC comparison with this chip. At least for the above posted clocks.
CryptiK - The key is hoping and expecting a crappy batch, then you get the golden ;)
Zoson - Looking forward to posting in your thread sir :)
So I guess you guys have seen this then ?
http://vr-zone.com/articles/first-lo...ck-/13795.html
Of course we did. But if you have a high end 1366 setup like _everyone_ in this thread... 2011 is not all that relevant. I suspect many people with R3E's will skip 2011 entirely(I am planning to). 11% IPC in best case scenarios is just not enough to get me excited.
Same here.. I'm sticking with 1366.
Does anyone know if R3E supports X56xx series CPUs like X5650?
I think it should spool, as long as the bios already supports the multiplier of that cpu. It's not official, but there are many reports of people using xeon 5xxx series cpu's on r3e. You can see them if you google it.
Thanks for the reply Zoson. Yea I have seen screenshots also but can't figure out why my board won't run a X5679 I have. Maybe because it is an obscure CPU?
It keeps giving me a Bad Bios Checksum error. And then it continuously searches the removable media and cd-rom drives for what I suspect a Bios file for recovery. Don't know if it's a broken CPU or lack of Bios support.
Maybe the daily OC could persuade some to jump the 2011 wagon...
Guys Its been awile I was wondering if anyone could tell me what the best bios is these days for the R3E I have been running some beta 0096 bios for awile now
It looks like X5679 is an i7 970. I would expect this CPU to work. It seems silly for me to ask this, but you cleared CMOS? Also, did you try 1401 bios?Quote:
Originally Posted by spoolindsm127
I'm not expecting 2011 to do any better OC than 1155. Same architecture + more cores = more heat and more volts required for similar clockspeeds.Quote:
Originally Posted by Leeghoofd
I think the general consensus is 0003, or 1401 are the best now. 0003 is the 'memory' bios. if you have memory that doesn't behave nicely with the r3e, 0003 saves you. 1401 is the same as 1301, but with additional CPU support.Quote:
Originally Posted by thobel
I'm going to need to test out 0003 I had an odd issue yesterday. I was asked to do some benches for a review running stock speed with Ram @ 2133 whenever I went stock cpu and 2133 ram i would only see 2 or 4gb of ram as soon as I went back to normal overclock all my ram was back and fine.
ps anyone have a link to the 0003 bios?
So got 0003 same issue Windows detects 4GB (2x2) however CPU'z see's all the memory (2gbx3) anyone have any idea's?
2011 socket seem best in OC,Quote:
Originally Posted by zoson
http://en.ocworkbench.com/tech/intel...g-5ghz-on-air/Quote:
Today, we just witnessed an overclocked LGA2011 processor. The processor runs on the Intel X79 Express chipset mainboard with a regular heatsink. We observe the temperature at IDLE, it reads around 45 degrees.
A check on the CPU monitor shows that the CPU is running at a multiplier of 41X with CPU CLK set to 120MHz. This gives us a reading of 4.92GHz. The memory was running DDR3-2400 at CAS 10. CPU Vcore was 1.51v.
From the readings, it looks like the new Core i7 would be a pretty good overclocker. Who cares about LN2 when you can achieve such high speeds even on AIR. The retail processor should do better
Adjust your tRL by +1 on all memory channels. That is the memory feature that 0003 has which all others do not. If +1 doesn't work, try +2.Quote:
Originally Posted by thobel
I don't buy hype. It's one thing to get into windows at 4.9ghz, it's entirely another to be fully 24/7(linx) stable.Quote:
Originally Posted by tistou77
New board is looking good :)
4.5GHz 155.5x29
1866MHz/3732MHz
7-7-7-20-1T-60
Bios 1208
LLC 50% / QPI LLC Off
Vcore (real_load) 1.323v
QPI (real_load) 1.279v
DRAM (real_load) 1.642v
http://img407.imageshack.us/img407/2...ghzlinx20s.jpg
http://img545.imageshack.us/img545/2...ghzlinx20s.jpg
89.6842 GFlops / 4.50931 GHz = 19.89 GFlops/GHz
I just set it at the closest settings as the OC run and it passed. I still have to fine tune and hopefully be able to lower voltages.
Holy crap! That is one hell of a chip man :eek:Quote:
Originally Posted by spoolindsm127
I wish my 990x could do that good. Oh well.. I sold it anyway.
That and quad channel memory. I have 8x2GB Hypers waiting for SB-E. :DQuote:
Originally Posted by Leeghoofd
I hope it won't be as [sarcasm] fun [/sarcasm] as SB, memory wise.
Very nice Spoolin :D
Thanks McDown and CryptiK.
It seems the same thing happened with the 980x/990X that happened with W3570/W3580. There weren't nearly as many Golden W3580s (if any) as there were Golden W3570s. Intel marketing gimmicks I think...
Finally a new bios is out for R3E though don't know what's new or changed in this one as it's not official on their website yet.
I hope it's not another crappy update that gives support for newer CPUs only like the previous release.
Hopefully it brings something worthwhile.
Rampage III Extreme 1502 BIOS
ftp://ftp.asus.com.tw/pub/asus/mb/LG...-ASUS-1502.zip
It appears to be slightly smaller in size as compared to the rest (except for 0704 and below), I wonder if they removed/replaced something or just coded something more efficiently?Quote:
Originally Posted by Sylver123
someone tested it? better than the 0003 or other about the voltages, OC, etc...?Quote:
Release Notes
--------------------------------------------------------------------------------
[ 1502 ]
----------------------
Supports over 2.2TB huge HDDs and RAID.
*Please install IRST 10.5.0.1026 when using this Bios version
I have no time to test it at the moment :(
there's a note below "Please install IRST 10.5.0.1026 when using this Bios version"..
what is IRST..?
there's a note below "Please install IRST 10.5.0.1026 when using this Bios version"..Quote:
Originally Posted by Sylver123
what is IRST..?
Wow, its "Intel Rapid Storage Technology", and that note means we probably got an upgrade to our ICH Sata controller firmware. I see no other reason they would tell us that, unless the RIIIE was getting a new feature that can only be taken advantage of via the new drivers that support it. Very interesting NOTE, lol...Quote:
Originally Posted by kaka_89
You are right.Quote:
Originally Posted by SkOrPn
The new BIOS has the new Intel OROM firmware inside.
Interesting... Has anyone tried the newer drivers, including any Alpha drivers such as version 11.0.0.1015 or higher yet? Also I tried to find info on OROM but did not find much. Not sure if Im searching properly or not but Id like to know what OROM is before I jump to it. Be great if this has performance improvement for raided SSD's.Quote:
Originally Posted by Brama
Any updates on the new firmware increasing performance on SSD's in raid0 ? ty
I just read that this bios needs to be matched to the IRST driver version, which means many of us will have to downgrade driver versions? I'm currently on 10.6, but this 1502 BIOS requires 10.5.0.1026 for OROM? I'm not sure I believe this myself but that's what I've been reading around the internet. How can a newer version not work?Quote:
Originally Posted by Chigurh
I cannot see 1502 BIOS needing 10.5.0.1026 IRST for OROM and no higher version or something will not
work or will be broken. I would imagine 10.5.0.1026 or higher IRST should be ok.
I have 1502 BIOS installed with two Crucial RealSSD C300 drives in RAID 0 on the Intel controller along
with IRST version 11.0.0.1015 Alpha and see no problems.
Quote:
Originally Posted by Sylver123
Me either really it's kind of silly that they mentioned that in this BIOS Release. Did you notice any gains in read/write? ty
Quote:
Originally Posted by SkOrPn
As i posted above it really doesn't make any sense. /shrug
Sorry, I haven't tested for any performance improvements.Quote:
Originally Posted by Chigurh
I absolutely agree and also do not see a reason why newer versions of IRST would not work properly. I can only assume that Asus made a mistake when typing that note out, when instead they should have said version 10.5 or higher. If it really does require that particular version, they then need to explain why in detail, lol... I for one use the newest drivers that I can find...Quote:
Originally Posted by Chigurh
From looking at the recent LGA 2011 reviews, it seems I'll be keeping my R3E for a lot longer now :D
I just wish I had a CPU good enough to take advantage of the board though, been stuck with my I7 920 that needs 1.38v for 4.2 Ghz for over 30 months now.
What is the chance of Gulftowns going down in price now that LGA 2011 is out?
Likely, smart buyers wouldn't be getting gulftown anymore. SB-E seems on the list, sadly I'll have to ditch my Lovely 980x and R3E ( bad clockers :/ )Quote:
Originally Posted by bhavv
I'll be testing that new bios out on the Agility 3's I have in raid-0
My OROM is currently 10.1.0.1008, Intel RST KIT and user interface is 10.6.0.1002
Wish I could afford to buy your lovely Gulftown, lol..Quote:
Originally Posted by Cookiesowns
I just updated yesterday morning to 1502 and the System report says the new OROM is 10.5.0.1034. Everything else is listed as 11.0.0.1015. Even though I'm running with the alpha drivers I see no indications of problems, system is running as smooth as ever.
Awesome man, I'll try that new bios once I get this project finished.Quote:
Originally Posted by SkOrPn
I currently have an offer of I believe 650-700 for my chip. I'm thinking I should just ditch my 2nd 580, mobo, and cpu all together to afford an upgrade to the SB-E with quad channel of ram, RIVE and cpu. If I don't sell the gpu, then I'll be stuck with the K, but I still think it's plenty enough.
But man, my own saying of "once you go xtreme you can't go back" applies here. The e-peen factor of the "X" designation is just so powerful hahaha!
lol, not sure I will ever afford an extreme, but the cheaper of the 6 core cpu's are looking mighty nice to me now... Maybe the 970x and another 6gb of memory is the best I could do. Might even throw two more C300's into the mix, lol, or sell them and get two Force GT's. Not sure whats better 4 x C300's or 2 x Force GT's...Quote:
Originally Posted by Cookiesowns
4xC300 in raid-0, or two sets of raid-0 or hell, even RAID-10 heh.Quote:
Originally Posted by SkOrPn
Unless you have a really good SATAIII controller like in the X79, the Force GT is pointless. C300's also have very good read/write when put under stress.
Yeah I know, that's why I own two of them. But the system still feels slower than my 4x Vertex drives in raid 0. I didn't want the Force GT for its sataIII transfers, I just wanted them for its possible 85k x 2, 4k iops... plus maybe I would eventually get a raid controller again...Quote:
Originally Posted by Cookiesowns
Well I did. RAID-0 OCZ Agility 2, under SATAII cable and Corsair SATAII hot-swap bay on my 800DQuote:
Originally Posted by Sylver123
Note that as with ssd's there should always be fluctuations but, take this with a grain of salt.. either way the read performance loss is a bit .. eh:
Test with the previous bios:
http://i.imgur.com/nT2hw.png?4871
Test with the latest bios:
http://i.imgur.com/55KZH.png
( was from a fresh boot into the array, didn't exactly let it idle for 2 minutes. )
Test is 2x100MB, 0FILL
I see. Well as with any SATAIII SSD, running them in SATAII mode has a bit of overhead, you'll loose a few notches on performance. What type are those Vertex drives? 2 or 3? Also, the ICH10R is incapable of going any higher than 750-800MB/s regardless of how many spindles/drives you have. That being said, if I had 4 SSD's at my disposal I would run them in raid-10 instead of 0 just for my own reasons. Less risk of data loss, as well as you may get better read performance. Though, I have this raid-10 8 spindle server, it performs like crap to even a single SSD. Though 3TB of raid protected storage is fun to play with, hahaha.Quote:
Originally Posted by SkOrPn
For your C300's your running on the ICH10r right?
Quote:
Originally Posted by SkOrPn
Vertex 2 or Vertex 3? As with any SATAIII drive, theres always some overhead with a SATAIII interface when running in SATAII mode. Plus, the ICH10R is only capable of doing about 750MB/s sequential. I haven't seen any higher results on itQuote:
Originally Posted by SkOrPn
That being said, if I had 4 SSD's 120GB up, I would raid-10 them :)
nobody has tested the new bios on performance, stress, voltage, etc...?
1502 is DESTRUCTIVE to large raid 5 arrays. I restarted, flashed the new bios, and two of my three raid 5 disks were suddenly listed as non member disks.
I used the 'plug them back into their original build ports and run testdisk' fix to get my data back. And now I'm waiting the 3 days it takes to re-initialize raid 5 parity on ich10r...
Yeah it appears the new bios did not give any gains in performance on raid. Wish I had thought of doing a before test with as ssd. I have 2 C300's using firmware 0007, and I have 4 of the original barefoot based Vertex's on firmware 1.6 I think it is. I do not have the Vertex's connected to my main system anymore, they are for playing with on my older systems here or for transferring files. One is installed in my EEEPC, just to give it that extra bit of perf. lol
Well, let's see how it runs in a 24/7 setup. I don't care about loss of 2-5% performance on 2 ssd's, I seriously don't care. What I care about is reliability and if the sleep issue have been fixed :)Quote:
Originally Posted by SkOrPn
My computer sleeps great, I just turn it off when Im done with it. 24/7 with these C300's used to be just fine when my computer was producing bitcoins last year, but now I just use the computer for what ever, mostly IRC, browsing and flashing smartphones and also a few VM guest installs. I also game now and then, but much more rarely these days. Like you said, we need more time with the new bios to know how 24/7 will work out. Maybe I should test my sleep behavior as I've never put "this" computer to sleep before. Just curious, what sort of problems were reported regarding sleep?Quote:
Originally Posted by Cookiesowns
I've had a few different ones. The main one I've been having is when the computer goes to sleep, and you press the powerbutton or wake it up, It will show the windows login screen, but then will hang there with no disk activity, the second one is that The system sometimes won't even go to sleep, just hangs with a blank screen.Quote:
Originally Posted by SkOrPn
I haven't gotten waking up from sleep problem in awhile, but it still happens on occasion. It really pisses me off, since right after I reformat I always forget to disable sleep, and the computer ends up sleeping then bsoding since it doesn't wake up, thus ruining my day of a fresh install :P
Yeah I can see how that would upset someone, I know I would not be happy. Since Im in raid I set my power settings to high performance and set the HDD to always on, never sleep. Then once a week or month or so I log out and hope GC does its thing overnight. Using the sleep button on my Keyboard just now seemed to work just fine, and then hitting the space bar woke it up without any problem, in fact I might use that more often because coming out of sleep was super fast. I wonder, in this "Sleep Mode" is there still power at the SSD's? If so, this might be even better than "Logging Out" for proper GC.Quote:
Originally Posted by Cookiesowns
Anyway, currently busy trying to update one of my Vertex drives to 1.7 non-destructive on a EeePC 1005HA finally.
I see. Are you overclocked? And what bios version are you currently on?Quote:
Originally Posted by SkOrPn
Today NOT overclocked, but tomorrow might be a different story, lol... I just flashed bios 1502 several days ago. Hey I just noticed Alpha IRST package 11.5.0.1109 was just posted. Interesting indeed... :)Quote:
Originally Posted by Cookiesowns
Wow everyone, major good news for anyone using raid on Intel RST. I was just reading Intels release info and particularly the NOTE's section on their latest alpha release when I noticed note #2 states and let me quote below in BOLD:
Quote:
Notes:
1. Known Issue is defined as a potential RST issue that has been replicated internally by the Rapid Storage Technology team but has not been root caused to be an RST defect.
2. This release will not enable the TRIM on RAID0 feature, but it will be added in the next RST 11.5 release. :woot: Contact your RST AE representative with questions.
3. This is the first release of RST that includes UEFI driver support. Note that the UEFI driver will not work with any previous releases of the Intel® Matrix Storage Manager/Rapid Storage Technology product. This means if you plan to use the RST 11.5 UEFI driver then you must use the RST 11.5 driver as well.
4. The RAID OROM version for this release is 11.5.0.1111, the driver and user interface version is 11.5.0.1109.
5. This release contains the CLI (Command Line Interface) replacement utility for RAIDCFG32. There is a 32 and 64 bit CLI utility which reside in the CLI 32 & 64 bit Utility component of this kit.
6. The Windows 8 version used during RST validation of this release is 8028.
7. When an ODD is in the Zero Power state, some ODDs may disappear when a hardware re-scan is initiated via the Windows Device Manager. A second hardware re-scan conducted before the Zero Power Idle Timeout value is reached (default is 60s) will return the drive to an OS visible state. This behavior is believed to be related to the length of time an ODD requires to become ready during power on from the zero-power state.
8. Microsoft* .NET Framework 3.0 is required to be installed on the system in order for the Intel® Rapid Storage Technology User Interface to install. Note that not all Microsoft* Operating Systems have .NET 3.0 installed by default. Refer to the Microsoft* website for details.
I almost don't want to believe it, lol...
You still need the Intel OROM ( raid controller firmware ) installed for raid-0 TRIM to work, but that's awesome news regardless. Another good reason why I should upgrade to X79 heheheh.Quote:
Originally Posted by SkOrPn
With that being said,
Has anyone noticed slower transfer speeds? My 1.5TB black is only capable of reading 75MB/s now. Quite interesting. There's less stuttering, as well as boot times "may" be alittle bit faster.
Well crap, they did not mention that after the note, as they usually do. Normally they state "This new TRIM feature is only possible with the newer OROM version" Like they did with the UEFI note below it (then again this alpha is not enabling TRIM either, so maybe it will state that in the notes for next release). So I just assumed lol... So, I would now assume we will never see TRIM on this board because I doubt RIIIE will ever get a OROM 11.5 bios update. Yeah another reason to upgrade for sure. Quad channel, EFI Bios support, Built in BT, and now TRIM is coming for RAID0? They sure make it hard to skip generations these days...Quote:
Originally Posted by Cookiesowns
Have not done any transfer tests, but I did notice all my SSD's are now performing benches much less than before, or so it seems. But, on a side note, everything seems to be very smooth and snappy.
Yeah everything things just as smooth or smoother, but the transfer rates sure have dropped.Quote:
Originally Posted by SkOrPn
Man, I really want to upgrade now, should have just sold the cpu and mobo to the guy instead of waiting.
WARNING: This is just intended as a comparison between old alpha drivers 11.0's to 11.5's. I'n my opinion this post is worthless, but I wanted to see if this latest 11.5 alphas did in fact drop some performance, but it looks like it did NOT. At least on single small barefoot SSD's it did not. RAID could be an entirely different story...
I just did a few more tests on freshly wiped Vertex's and found nothing notably different from Alpha 11.0 to 11.5. Still wish I had done some recent b4's on Intel's official 10.6 release. I have 3 original Vertex's (all purchased the same day they were released for sale), and one brand new one (RMA replacement several months back). The wear should be similar on the old Vertex's. The new one has not been used "much" yet. This goes to show that the rma replacements, although physically look better are coming with crappier nand lol. All four tests were conducted on BIOS 1502.
Top left is the brand new Vertex (non-turbo) flashed to FW1.7 on ICH using the RST 11.0 drivers. Top right is the old original used Vertex flashed to FW1.7T, using the same Alpha 11.0 drivers. BOTH drives formatted to FAT32.
IRST 11.0
http://i.imgur.com/Nb1MTl.png http://i.imgur.com/uXU05l.png
Bottom left is the same old Vertex flashed to Turbo FW1.7T on ICH using IRST 11.5's and formatted to FAT32. Bottom right is the same drive formatted to NTFS on ICH also using IRST 11.5 drivers, just for s&g. Was curious if there was a performance difference on such small SSD's, guess there is not. Since I have them attached to USB 3.0 adapters, maybe I should test them formatted to exFAT???
IRST 11.5
http://i.imgur.com/AIEjdl.png http://i.imgur.com/pb8Bcl.png
IRST 11.5's appear to have given me a really super-slight performance increase. NO?
I have to admit, I did NOT test 11.5 before reporting my previous statement of slower benches. I had only had benches from the 11.0 alpha release (which I thought would be identical to 11.5). So, your findings with slower transfers kinda hit a bad spot with me and I wanted to make absolutely sure this was the actual case. Well after doing two more tests on the new 11.5's, I have to admit I do not see a performance decrease, at least not on these old SSD's of mine. I have NOT tested any of my storage F3's or my Raptors for that matter, but it looks to me like 11.5 is Ok in regards to single drive perf. If only I was willing to drop back to 10.6, we would know if the alphas are a step back or not in performance. Unfortunately I forgot to take a screen shot when I tested my raid array on IRST 11.0. Otherwise I would have done a comparison with it as well.Quote:
Originally Posted by Cookiesowns
It's not the drivers that's slowing us down, it's the new OROM included in the latest Asus BIOS. I did my own tests with the same INTEL RST driver and install, look at the difference.Quote:
Originally Posted by SkOrPn
:) I have a nit picky thing of matching drivers and firmware version. Seems like my old combo gave me the best performance.
I had a feeling that was the case, in fact I was thinking because 1502 now enables over 2.2TB drives, that maybe somehow what ever they did was the cause of the slow down. All my results I am getting with 1502 seem not as good as when I was on 1401. If it was important enough I would flash back to 1401 and do a few benches, but I don't believe it it is. lolQuote:
Originally Posted by Cookiesowns
The only bios flash I care to do at this time is one that turns this board into a x79 R4E ;)
I have also noticed slower SSD performance with my two Crucial RealSSD C300 256GB drives in RAID0 since installing the new RST 11.5.0.1109 Alpha.
I reverted back to RST 11.0.0.1015 Alpha I was using before and my performance is now back to normal again.
Hmm, very interesting. This is going to force me to do more tests, lol... Thanks for the heads up Sylver123 about your C300's...Quote:
Originally Posted by Sylver123
Today I have tested system on a socket 1155.
I have been pleasantly surprised by its productivity surpassing my present system.
Now I intend to sell my 1366-motherboard and processor (Rampage 3 Extreme and I7-920-D0).
Wanted to thank all participants of the given theme - your councils were it is very useful.
Good luck to you all!
:)