thats what i figured, but i thought that there were more entries than that, guess not......now that everything is on linux i guess i can get some good number
![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
-
Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.
CPU OC with SMP/Bigadv TPF & PPD Database
- Thread starter jebo_4jc
- Start date
thats what i figured, but i thought that there were more entries than that, guess not......now that everything is on linux i guess i can get some good number
thats what i figured, but i thought that there were more entries than that, guess not......now that everything is on linux i guess i can get some good number
All the data looks like its there...I just took a look and there was a filter on the spreadsheet, so not everything was showing.
excellent. thanks for fixing that. i noticed that filter icon but didnt know how to turn it off
sirmonkey1985
[H]ard|DCer of the Month - July 2010
- Joined
- Sep 13, 2008
- Messages
- 22,412
whats messy about it.. looks fine to me.
Yeah, someone cleaned it up.
will do later...
I'm locking the spreadsheet for editing for now. I've seen two issues:
1. We're missing some data. There were a couple results for i5 2500k that are now missing. I have those results saved to an archived copy from 4/4. I am going to check to see if any other results are missing.
2. I've seen some "filters" being applied to the data. This is fine for those who know how to use it, but the people/person that keeps doing it leaves the filter applied, and the data looks like it disappears for those who don't know how to remove the filter.
1. We're missing some data. There were a couple results for i5 2500k that are now missing. I have those results saved to an archived copy from 4/4. I am going to check to see if any other results are missing.
2. I've seen some "filters" being applied to the data. This is fine for those who know how to use it, but the people/person that keeps doing it leaves the filter applied, and the data looks like it disappears for those who don't know how to remove the filter.
everyone can PM me their data
yeah just more strain on you though...
Bump , just added another 16 points of data for a 2600k
Mainly SMP, but a few more big adv numbers as well...should give people a good idea of what it is able to do on SMP
Very interesting, on one unit it got a min time of 1:24 .... which was 70k ppd but that one had an average of 2:00 flat
Mainly SMP, but a few more big adv numbers as well...should give people a good idea of what it is able to do on SMP
Very interesting, on one unit it got a min time of 1:24 .... which was 70k ppd
but that one had an average of 2:00 flatwill start playing with it on thursday...
Untitledone
[H]ard|DCer of the Month - April 2012
- Joined
- Feb 28, 2011
- Messages
- 1,486
Just added my weak sauce scores to the SMP TPF Data base! Have fun with the pile that I added. It will let you know what the crappier quad cores can do.
Untitledone
[H]ard|DCer of the Month - April 2012
- Joined
- Feb 28, 2011
- Messages
- 1,486
Sorry for the double post but it looks like the clock speed column got screwed up on the data base. Maybe its just my computer, but the clock speeds for my processor got set to around 4ghz and I edited them back to what they were supposed to be. I just checked and now a bunch of other people's got changed to mine. So if someone else who started the page can check it out and see if im crazy or not.
confirmed, some one fucked up alot of the clock speeds, all my 2600k numbers were at minimum 2 GHz off(on the low side)
Untitledone
[H]ard|DCer of the Month - April 2012
- Joined
- Feb 28, 2011
- Messages
- 1,486
When I went in there it looked like some random drags happed. Most of my clock speeds got set to processors that where near by. I went and fixed mine by editing one and then dragging it down to the rest of mine. Everything seemed alright, all of mine were fixed and I made certain that I did not edit anyone else's. However when I checked later and noticed that ALL of the clock speeds were edited on ALL of the porcessors I knew something was wrong. Maybe there is a bug? I looked at all the processors that got set to my clock speed and it appeared that there were around double. So maybe it is a bug that when you drag the box to copy it is not actually applying it to the boxes you think they are. Only reason I think its a bug is because only the clock speeds got messed up. If someone really wanted to screw with it they could have messed with everything.
Untitledone
[H]ard|DCer of the Month - April 2012
- Joined
- Feb 28, 2011
- Messages
- 1,486
Well, I am glad that I am not the only one that can see it. I was going crazy thinking that I screwed everyone's data up. Hopefully with the new database and logger it can get automated so that we dont have all this manual stuff.
The new one is unrelated to the old one and it looks like the clock speeds make sense:
Pulled the 138 results in the current bench database into Excel and the frequencies look reasonable. All data in the current bench is automatically reported, but there are some gaps with older versions of the benchmark that did not capture as much data.
BTW if anyone is good at PHP/ MySQL let me know because we probably need a better front-end to this. May be able to hook you up with some folding hardware for efforts.
Untitledone
[H]ard|DCer of the Month - April 2012
- Joined
- Feb 28, 2011
- Messages
- 1,486
That screen shot you posted looks reasonable. We definitely know someone messed up the old one though because we checked our own entries and they were messed up.
I would be happy to run the new one once you get it ready. I don't mind doing some beta testing but I am not a coder or anything so I would only be able to forward errors to you guys and wont be able to troubleshoot much.
I would be happy to run the new one once you get it ready. I don't mind doing some beta testing but I am not a coder or anything so I would only be able to forward errors to you guys and wont be able to troubleshoot much.
your clock speeds for a 2600k are wrong, i had mine in the database as 4700, not 4626
just a heads up, some data may be skewed there as well. i'm assuming mine is the one reported at 4626 or 4190, due to me having put up ~6 results or so, if not more for the 2600k.
just a heads up, some data may be skewed there as well. i'm assuming mine is the one reported at 4626 or 4190, due to me having put up ~6 results or so, if not more for the 2600k.
I assume you're talking about the google spreadsheet? I've not touched it. The results in the SQL DB are purely from the bench application - no manual entries, and no import from the spreadsheet.
H.
New version will be up tonight - just have to fix the version string, and update a couple of cosmetic features. Don't expect anyone to be coding anything - I'm happy to do that - just need any error reports.
H.
The clockspeeds PJ posted are from the SQL database - for 2600K's its mine at 4160, Jebo's at 4500 and PJ's at 4626.
H.
And just to note, that screenshot is just a quick pivot table of the results in the MySQL DB. Haitch has worked diligently to capture a lot of information (~60 fields.) The other cool thing is that you really just need to enter your user name, path to the F@H directory, and then "bench.exe -live" and you can continuously report results.
Nope, a big design principle has been gathering data automatically. User reported data can suffer from human error in both accuracy and completeness. Also, some people do not know how to properly sort/ maintain information in spreadsheets (as the Docs version has experienced) and thus by keeping the DB and letting users probe results through other methods, we help keep data integrity a bit higher. Haitch's *awesome* work is something to check out.
Untitledone
[H]ard|DCer of the Month - April 2012
- Joined
- Feb 28, 2011
- Messages
- 1,486
Now that is what I am talking about! You just need someone to create a nice and clean front end with some useful features. Some suggestions are; Sort by PPD, Sort by Clock speed, core count, thread count, User name, WU. Pretty much like the columns in the spreadsheet. I pretty much have used those to sort, and get an Idea of how various processors perform.
Maybe add a check box in the program so the user can set if the computer is dedicated or not. This would allow us to get better results on dedicated machines, and also machines that get everyday use.
The front end I'll look at, but web is not my fortè. We suspect most people will prefer to suck the DB info into Excel and do their own analysis.
As for dedicated vs not, there a user notes field where that can be specified, additionally by looking at the std. deviation of the frame times you can get a good indication - my dedicated rigs have a std dev < 1 sec tpf, my main working machine has std devs of 21 - 130 secs, and others have even higher.
H.