Gautam
-
Posts
382 -
Joined
-
Last visited
Content Type
Profiles
Forums
Events
Blogs
Posts posted by Gautam
-
-
I bet that he will not come.....
Reminds me of "another" person doing such things without providing real evidence NEVER till now....
Hey George, how much would you pay for a set of DDR3 that does 1040 7-6-6-15 for 32M, and a board that does 650 FSB along with it? Shall we get the bidding started?
-
I don't have a strong opinion on this...yet...as I said, for the clocks all the scores look about right to me...I'm with hipro though, if he can reproduce similar overclocks live, it'll ease our minds and put the topic to rest.
-
And the board and memory too
-
You know, I'd be all for a third party confirming that that board and memory combination can indeed do 650MHz 5:8 7-6-6-15 for 32M. (Even if it's at a low CPU speed)
-
I mean that no one has ever done such high memory speeds even at CAS 8 or whatever on the Evolution, let alone CAS 7. How are you doing such speeds at only 1.61v with a clearly stock cooled board according to your pics?
Vince has P1 and P2 off, with LN2 on the northbridge. Jody I believe did 600 flat for Everest with them on, also with LN2 on the northbridge, or single stage for his latest run.
As for 8800 GTS, I see no one even coming close:
http://www.hwbot.org/quickSearch.do?hardwareId=GPU_1255&name=GeForce+8800+GTS+512+Mb+%283974%29
elmor and no_name are the only ones pushing mid 1100's. Even 1100 is a struggle with this card normally. And somehow you got TWO that are 100MHz better than the best on that list.
We can go through this hardware by hardware, but you have had so much highly unusual hardware that it's clearly going to take a while.
I'm not calling you names or suggesting anything, just making some observations. And you're right, this is going to go nowhere. Again your scores theoretically don't seem impossible or anything, but when you sit down and look at all of these anomalies together it raises an eyebrow...you have gone through a lot of insanely cherry hardware for a guy that claims that none of it is even handpicked.
-
Well, I admit that I have spent a lot of time handpicking my CPU at the warehouse I used to work at, but I don't think I have the best one (boble and andre had 6.8G chips, jowi has 6.53G 32M "without pushing it much" chip)
No one has come close to 6.7GHz for 1M.
the scores are irrelevant to the memory kit I was using. We could hit roughly same clocks with two other DDR3 sets I currently own. (those are not handpicked in any way)You must be the only one then, that speed at tRAS 15 is uncompletely unmatched.
No. I've seen multiple D9GMH sets that could do 600MHz CL4 32M at 2.35V DMM or lower (mine can't do that)I myself had a pair that could do that. It still needed 2.8v for 660 4-4-4 1M, and that at 2:3 divider. 1:1 is MUCH harder.
Why do you think it's best? I reckon mine's one of the worst
Anyone, and I DO MEEN ANYONE who had put mch subzero could hit 650+ benchable.
Fine, I'll give you that.
And why didn't you address the board? That's the strangest thing here. An X48 that can do 1040MHz mem at ANY timings on 5:8 at 1.61v??? And of all boards, the P5E64 Evo which is especially weak in clocking memory??
At least understand why your results fall under scrutiny every time. They are always very very unusual in fashion. You always use strange hardware configurations and manage to hit unheard of speeds with it. (And always at efficiencies that no one else can match)
(e.g. Blitz Extreme at 610 PL8, 790i at 616 with P1/P2, 8800GTS at 1240 core, and let's not forget the magical 8MB cache E6400)
And now all of this. Maybe one aspect of any one score of yours might be possible to believe, but all of this coming from one person, time and time again?
I don't think you've ever once shown a "normal" benchmark score. Every score you've ever put up in the past two years has always had something very odd about it. No one else has ever done any scores that are even similar to the ones you do in nature.
-
You seem to be an extremely lucky guy.
You have the best CPU in the world by far.
You have the best DDR3 in the world by far. (tRAS 15 at 1040 is unheard of. I'll admit I've done similar speeds myself, but certainly not at such low tRAS...and completing a 32M run at over 2.3v??)
You have the best DDR2 in the world by far. (660 CAS 4 is not that difficult with the right pair, but at 1:1???)
You have the best P5E64 WS Evolution in the world by far (while good in its own right, this board behaves nothing like a Rampage at all. The FSB isn't so much an issue as it is the ram clocking, and not only that, anyone that's tried this on multiple X48's can attest that the 5:8 is extremely difficult at anything past ~620 normally, even if the ram is up to it. One would need well over 1.7v MCH to have a prayer of pulling off such a ram frequency, and even more still to get 5:8 working...)
You have the best Biostar T45 in the world
All of your scores seem plausible to me. What's so hard to stomach here is that you apparently had lightning strike four times in the same place.
-
Well said Jason, there is way too much of this going on. (I don't even need to name names)
-
When I look at the poll, I see that the option with the highest amount of votes is adding performance with points.
-
Yeah, it always says that IIRC. A compare link would settle it for certain though.
-
It's obvious he ran it at 1024x768 without meaning to. That's what 06 by default will do if you have a monitor that doesn't work at 1280x1024.
Either way the score isn't valid.
-
Add all 3...no reason not to. $20 for the advanced is not a big deal. The presets are there to shift the bottleneck from CPU to GPU which is a nice touch. Different rigs will do better on different presets, so it makes sense to award points to all 3. It is like running 3 different benchmarks.
The "not forcing people to buy" argument makes absolutely no sense, hwbot does nothing but force people to buy hardware as it is.
Scores always take money; that's just a part of life. -
Heh.
Even without any links you wouldn't need to have "cleaned" graphs. I still think that automatically generating the graphs would be fine. Outliers would still be easier to catch. I can't see any downside to having current results displayed graphically.
-
Yeah TBH in my ideal image, each dot links to the compare link for that score. That way if you'd see an outlier you could report it immediately.
But that might be pretty difficult to implement.
-
I love charts, and technically they're pretty easy to implement. The hard part is picking out 'good' samples. As hwbot receives 300 to 500 submissions a day, the result moderators barely have time whether essential info is correct (correct score, correct cpu/gpu and has validation), so there are quite a lot of scores which have wrongly matched cpu/gpu speeds.
There are multiple ways to solve this: more mods, better tools to detect these faulty scores for mods, or better algorithms for charts which detect bad results. We're working on the latter 2. For example, we just made the 'overclockability' chart better for videocards:
http://www.hwbot.org/quickSearch.do?hardwareId=GPU_337&name=Radeon+9800+Pro+%28498%29
(scroll down a bit). It will take some time before we are confident we can add more charts. Better 1 good chart than 10 poor ones.
What I was thinking was just having the charts created dynamically based off off of all submitted results. In fact, this would make reporting false scores even easier as outliers on a chart would get caught much quicker and more easily than if they were on a list. However the main benefit as I see it would just be in aiding in extrapolation and visually being able to see scaling.
Note for example the ever-so-slight curvature in the speed vs. time plot. Subtle, but its there, just like any good hyperbola.
: -
What's up guys.
I was chatting some SuperPi with a friend last night and figured this might be a neat thing for you guys to implement if you have any spare time. (what's that?)
As you can see from the graphs below that Vapor whipped up, time and CPU speed are inversely proportional. With a trendline like this, or even without, its easy to visually predict what sort of time you'd end up with at a given CPU speed. This would work equally well for SuperPi 32M and PiFast.
Whenever people ask how they're times and scores stack up, I always point em to hwbot. It'll be even reasier to see how one compares to others or what one should be expecting at certain clocks graphically like this.
Additionally perhaps we could consider displaying performance products for all scores and showing how someone's scores compare in efficiency to others.
I think this stuff should be fairly easy, for these graphs Vapor and I each spent a couple of hours collecting data points off the hwbot 1M ranking. Would be nice to see the same process automated.
Thanks again for all your continued hard work.
cheaters sucks..
in General overclocking
Posted
That sounds pretty good to me...if either of those two can replicate then there will be no doubts. You can explain to them however it is you did it, then they can show the rest of us results.
(Someone like Kinc who isn't involved in hwbot might be worth a consideration too)