Hyperhorn

Any new thoughts about "is L3 cache running at core or ring frequency" thing? What would you guys think if core AND ring frequency boost L3 cache performance at a 2/3 (core) and 1/3 (ring) relationship? Wouldn't it be a nice idea that L3-cache runs at core frequency and ringbus only boosts the interface between the cores and the L3-cache? Given L3-cache and ring share one clock domain in fact and run at the same frequency - for what reasons would the core frequency affect synthetic L3 performance values much more than the ring frequency or e. g. CPU NB clock frequency of AMD CPUs?

Very clean insulation and interesting results! How much Vdimm for the 5-9-5-20 @ 32M try?

Just imagine 6.2 GHz @ 1.2 V would be real. "Does it scale?" Not mentioning that info was part of the IDF 2013 slides I will grab me some popcorn and you can go on with the violent NDA protection. Pics or it didn't happen!

For now I have one result to add: 2006/06/30: 9.828s by Coolaler (link) As far as I know coolaler did the first sub 10s runs and this result is the best he achieved that day. Unfortunately the screenshot(s) are gone ... btw due to the list I already could sort out two false positives from 2005 and 2008 I (and most overclockers back then ) thought that were WRs. Edit: What happend with No_Name's 7.437s? The link in your list is broken and 7.440s is the best I can see from No_Name for the QX9650-category. However this thread at XS includes the screenshot and system details of the 7.437s WR result, so if the broken link can't get fixed this is a vaild source. (btw I have the modded P5K used for this WR at home actually as a collector's item. :banana:)

I had a good laugh when I read " One does not simply overclock a C0 CPU".

That's actually the brandnew wood tweak applied correctly. A Titan modded this way will not suffer from insufficient VRMs for extreme overclocking anymore. Also condensation is not a big issue which makes this solution superior to the usual zombie mods from all the modding noobs out there. World records incoming! *Knock on wood*

The current solution is the best way to encourage overclockers to stand clear of tactical backups and present the best achievable results right from the beginning. As Hwbot should be a place where folks can see full potential of hardware unleashed by overclockers and not only scores that are "good enough" regarding the competition, the status quo has its justification. Also usually when you compete in a category at a subsequent date you have access to more data, newer/more software (BIOS, drivers, tools, benchmarks, OS improvements etc.), sometimes even small hardware advantages (new revisions/batches, improved silicon quality etc.) and know exactly what scores you have to beat. Most Hwbot overclockers I know don't try very hard to get an improvement of 0.001 s in a benchmark on principle, but to beat an established result. When time goes on you get a natural advantage, but the current solution is a good way to encourage overclockers to submit their scores early (to get a possible advantage over the competition) and by a side-effect it balances the natural advantage that later overclockers have in very specific cases a little.

Not that many people I guess, but it's a great card - no doubt about that. I forgot I was talking about Fire Strike Extreme, so it's likely the same problem with 3-way-SLI and the combine test already described in the thread. That means it's not an isolated Titan problem, but a general driver issue.

My experience with Titan and 3DMark so far: 3-Way-SLI with Geforce GTX Titan results in unusual behavior. With the exact same driver settings (used Geforce 314.09) the framerate drops from 14.8 Fps (2 x GPU) to around 5.5 Fps and the screen "twitches" already when loading the combined test. Also while Game test 1 scales from 2 to 3 GPUs (ca. +25 %), I noticed the framerate in Game test 2 is reduced by about 10 %. The GPUs were not overclocked and the same goes for the Core i7-3960X used fot the few runs. 2 GPUs vs. 3 GPUs (one dedicated for PhysX @ driver) improved the overall score by around 10 %. 3-Way-SLI results in lower scores because of the bugged combined test.

I don't know. I've seen the video you uploaded and der8auer posted and it is fine in my book (but doesn't include a benchmark). If such a video exists it would be useful to include it here. btw to point that out: I'm not calling for actions here, rather I wanted to share my point of view on this topic as an interested observer.

Actually it's "only" 166 MHz better than duck`s Netburst record, so I wouldn't dramatize that by blowing the numbers up. ;p Remember that Andre Yang is the CPU-Z record holder with a gap of 292 Mhz at the moment and I think nobody questions his score seriously. (Yes, Zambezi is quite new and you can't compare the values directly, but what I want to say is that it is in fact possible to achieve legit records with quite a gap.) Some ideas to prove the legitimacy: Trusted overclockers who managed to achieve 8 GHz numbers with Netburst-CPUs in the past could get the chance to work with the same processor(s). Benchmarks like Super Pi 1M from start to finish together with CPU-Z on a video also would be an indicator of the supremacy of wytiwx`s CPUs. Did wytiwx tell accidentally how many Netburst CPUs he has access to? If there are hundreds of Netburst chips from a fresh source (not Ebay, maybe a Chinese sale platform or bankruptcy items) I don't think it's that unlikely that a very limited amount of them is good for the clock frequencies we see now. It's a difficult situation: One the one hand a proper CPU-Z validation fullfils all needs requirements for a legit Hwbot score on the other hand it's surprisingly one overclocker seems to exceed the limit (or what we thought it is) of an exhausted architecture so easily with several CPUs and we remember that the way CPU-Z clock readings work were changed when Windows 8 were released at the end of 2012, what led to incorrect values in some cases not a long time ago.

The PC version of the new 3DMark can be downloaded now from one of the download mirrors or Steam and I'm honestly pretty amazed one of my work mates managed to test the performance of 22 graphic cards in the test scenes Cloud Gate, Fire Strike Extreme and Ice Storm from 720p with no AA/AF up to 1600p with 8xSGSSAA 16:1 HQ-AF on day 1. If you're interested how HD 7970 (GE), HD 7950 (Boost), GTX 680, GTX 670, GTX 580 etc. perform please have a look at the results from PC Games Hardware: - German version: http://www.pcgameshardware.de/3DMark-Software-122260/Tests/3DMark-im-Test-Benchmarks-Download-1047157/ - English version: http://translate.google.de/translate?sl=de&tl=en&js=n&prev=_t&hl=de&ie=UTF-8&eotf=1&u=http%3A%2F%2Fwww.pcgameshardware.de%2F3DMark-Software-122260%2FTests%2F3DMark-im-Test-Benchmarks-Download-1047157%2F (Note: You can switch between the results by clicking the "tabs" at the head of the benchmark result box. As usual high quality driver settings were enabled, so you will likely get higher results with your default bench-settings.) If you look at the results it's easy to see why AMD suggested Fire Strike as a good test scene in their press release. Also here's the official Futuremark trailer released today: What results have you got and what do you think about the new 3DMark from Futuremark? Please also post links to other GPU/CPU benchmarks results you find on the internet, so we get more data to get an idea about the scaling, what scores can be considered normal etc.!

Achievement unlocked: One year closer to the grave ... oh I mean All the best and enjoy your birthday!

Please note that the DDR(3)-rating differs from the actual I/O clock frequency in MHz shown by CPU-Z. So 1,117.8 MHz conforms DDR3-2236. 2,234 MHz as in the title would be DDR3-4468, so a Hwbot mod should fix this.

I like that it says "FAITHFULL" at the bend, so go on. By the way I think this was also a not so common LN2-setup with the sock insulation tweak and cooling tuned by Mc Donald's: http://hwbot.org/submission/2334154_pixy_3dmark03_2x_geforce_8800_ultra_108717_marks

Simply amazing! You have beaten the old Intel/Netburst/Cedar Mill record (by duck) by almost 80 MHz. Today Bulldozer CPUs seem to be the easier way to get 8 GHz results, so this one is a real gem in my eyes. :ws:

So close ... Only 8.75 GHz left with this spectacular FX-8350 to get the CPU-Z WR!

Now we get new 32M WRs on a daily basis delivered by Andre Yang. The CPU frequency is very impressive. A CPU frequency of over 7.15 GHz is needed for that (rather 7.2 GHz looking at the average efficiency in the 3770K top 20), so I'm not sure we will ever see sub5 with current Ivy Bridge CPUs. But who knows what happens when a very experienced SPi tweaker gets a CPU with a good IMC that is capable of over 7.1 GHz @ 1M and has excellent binned RAM at hand.

Team timeline is a positive change, I think (bringing back) an user timeline could have a positive effect as well. I like how positive Hwbot developed recently and can't deny a certain positivity here.

Brand: Asrock Model: HM55-MXM Validation: http://valid.canardpc.com/show_oc.php?id=2402417 --- By the way there's another similar thread, which were created later: http://hwbot.org/forum/showthread.php?t=50629 I think one big thread with the same topic is enough ... (In case Hwbot staff prefers support tickets in general likely both threads should be closed, but of course that's up to you.)

Just amazing, well done!

Great result! Nick, what is the real Vcore you applied? 2.xx Volt?

Sock tweak enabled and cooling tuned by Mc Donald's Congratulations, nice score!

What a coincidence, just yesterday I've gone through the SDRAM ranking and thought it's about time that somebody challenges the record holder. I'm curious what you can achieve with LN2. Great result! :ws: btw it would be nice if you could tell us which ICs were used for this module. I didn't expect to see Hyundai ICs at the top SDRAM spots to be honest, but I'm not a SDRAM expert ...

Wow, you could use the DDR3-2933-divider with air-cooling. That's what was holding me back. Such a CPU is pretty rare: http://hwbot.org/forum/showthread.php?t=52322 New national and team record! :celebration: