sumonpathak Posted January 13, 2012 Posted January 13, 2012 ^^^look closely....you better rephrase it (am a law student...so always try to find loopholes ) Quote
Massman Posted January 13, 2012 Author Posted January 13, 2012 I re-read and still don't see the issue. 1) We don't have to care about how many teams participate => we can make an old-school competition with 5 participants, no problem. 2) Prefer 'fun 20-team' contest over 'boring 30-team' contest => as long as it's fun, we don't have to care about the quantity of participating teams. So, we can have as much legacy stages as we want. If there's only 5 teams competiting, that's fine for us. As long as it's fun for those participating, it's all good. Quote
Luebke Posted January 13, 2012 Posted January 13, 2012 you could maybe make two separate cups: one for the sandy-e/7970-pros and a classic-cup wich uses a lot of different hardware from different manufactures i think competitions with a lot of different hardware are much more fun and (to me) much more interesting... for example: whats better in aquanox/3d-mark 99: a dx6-card from nvidia, ati, s3, sis, kyro, voodoo, 3d-labs...? i don´t know but might be fun to find out in a competition Quote
IanCutress Posted January 14, 2012 Posted January 14, 2012 There are 10080 minutes in a week. Each minute: 1st gets 2pts 2nd gets 1pt 3rd gets 0.5pt Then at the end ot the week, 1st gets 10000 pts 2nd gets 5000 pts 3rd gets 2500 pts ... So if you go from number 1 the whole week to 2nd in the last few hours due to a sandbagger, you still get a win. If you've saved it until the end, more fool you. Quote
I.nfraR.ed Posted January 15, 2012 Posted January 15, 2012 (edited) And what if I'm busy during the week, because of my job and bench only in the weekend, producing the score at the end of Sunday. Does it mean I'm a sandbagger? Not that I'm a factor in the OC competitions, but almost all of my scores for competitions are done this way. Always at the very last hours, due to lack of free time. Others are done in the beginning of the comp and then I had no time to improve. Since it's all about fun (no?), I'm not sure it's good to give advantage to people with a lot of free time. Edited January 15, 2012 by I.nfraR.ed Quote
IanCutress Posted January 15, 2012 Posted January 15, 2012 No, that's just the nature of competitions. If an event started and ended on a thursday, you may put a score up over the weekend, but if someone beats you, you will (or others will) try and find the time on wednesday or thursday to rebench. Ultimately the person who has only one chance to bench probably wont win - if you only played tennis once a week, you wouldn't expect to win a title on the tour. Quote
Hondacity Posted January 15, 2012 Posted January 15, 2012 And what if I'm busy during the week, because of my job and bench only in the weekend, producing the score at the end of Sunday.Does it mean I'm a sandbagger? Not that I'm a factor in the OC competitions, but almost all of my scores for competitions are done this way. Always at the very last hours, due to lack of free time. Others are done in the beginning of the comp and then I had no time to improve. Since it's all about fun (no?), I'm not sure it's good to give advantage to people with a lot of free time. We are the same. I benched the last few days before the end of the competition. I don't sandbag, but I only bench when time permits. Quote
Massman Posted January 15, 2012 Author Posted January 15, 2012 you could maybe make two separate cups: one for the sandy-e/7970-pros and a classic-cup wich uses a lot of different hardware from different manufactures i think competitions with a lot of different hardware are much more fun and (to me) much more interesting... for example: whats better in aquanox/3d-mark 99: a dx6-card from nvidia, ati, s3, sis, kyro, voodoo, 3d-labs...? i don´t know but might be fun to find out in a competition Naaaah. There will only be one Team Cup and a large part will be with older hardware. No need to have another competition be all about the latest high-end hardware . There are 10080 minutes in a week. Each minute: 1st gets 2pts 2nd gets 1pt 3rd gets 0.5pt Then at the end ot the week, 1st gets 10000 pts 2nd gets 5000 pts 3rd gets 2500 pts ... So if you go from number 1 the whole week to 2nd in the last few hours due to a sandbagger, you still get a win. If you've saved it until the end, more fool you. I like the mathematical balance in this one. Good suggestion! Quote
I.nfraR.ed Posted January 15, 2012 Posted January 15, 2012 No, that's just the nature of competitions. If an event started and ended on a thursday, you may put a score up over the weekend, but if someone beats you, you will (or others will) try and find the time on wednesday or thursday to rebench. Ultimately the person who has only one chance to bench probably wont win - if you only played tennis once a week, you wouldn't expect to win a title on the tour. With the difference that we are not professionals. Ultimately the person who has only one chance to bench probably wont win So, where's the problem then? If he's so good that noone could beat him, it would be the same if he had the time to bench in the beginning and uploaded the score, or benched at the end of the week. But in the second case, the rules would artificially put him on 2nd place. Don't get me wrong, I still want some different competition rules that would make it more interesting. Quote
Massman Posted January 15, 2012 Author Posted January 15, 2012 Fyi, with almost no development time available until the beginning of April, I think we'll kick of the competition on May 1st. It'll be a 3 or 4-month competition with 10 stages. We're looking at hw requirements of maximum 27 different cpus, 7 different memory kits, 22 different gpus from 11 different vendors. About 55 scores. Quote
knopflerbruce Posted January 15, 2012 Posted January 15, 2012 If you've got thoughts about stages (benchmarks and HW limitations for example), feel free to post them here:) I guess the input here will be important when Massman makes the final decisions. Some stuff i posted in a thread not accessible to everyone:p 10 stages, that means we can use alot of different HW and benchmarks. I suggest about 50/50 2D and 3D, maybe with some HDD/MB/RAM benchmarks in addition (not more than one of each of the 3 types). Maybe 4 2D, 4 3D and 2 of the 3 others, depending on how well we know the different benchmarks (we don't want to use untested benchmarks in a serious competition). As for HW limitations, 50/50 AMD/Intel for 2D and 50/50 AMD/NVIDIA for 3D? Then you have something for everyone. Quote
IanCutress Posted January 15, 2012 Posted January 15, 2012 Aquamark3 on IGP, limited to DDR2. DDR3 low frequency. AS SSD with Ramdisk. wPrime long, but overall score is time*cpu threads. so 10 secs * 16 threads = score of 160. Quote
Massman Posted January 15, 2012 Author Posted January 15, 2012 AS SSD with Ramdisk. AS SSD Storage Benchmark + Software Ramdisk = new memory benchmark? Quote
Luebke Posted January 16, 2012 Posted January 16, 2012 @knopflerbruce: sounds good, but one thing i´d change: 50/50 AMD/Intel for 2D and 50/50 AMD/NVIDIA for 3D? i´d make one non intel/amd and one none amd/nvidia stage via/ibm/cyrix for 2D and sis/s3/3dfx/3dLabs/kyro for 3D Quote
saint19 Posted January 21, 2012 Posted January 21, 2012 Would be nice extreme cooling vs extreme cooling and iar or water vs air or water.... Quote
Luebke Posted January 23, 2012 Posted January 23, 2012 stages by type of cooling? interesting idea... might give a chance to good overclockers who don´t have dice/ln2/cascade... and the others could participate twice: once with "normal" cooling and once with extreme cooling. might be interessting to see the differences on same samples... Quote
reggiesanchez Posted January 24, 2012 Posted January 24, 2012 Would be nice extreme cooling vs extreme cooling and iar or water vs air or water.... Cooling limitations are a bad idea, just provides a platform to either lie about results or spend your day accusing someone of cheating. But love the idea of a big structured competition. Quote
saint19 Posted January 24, 2012 Posted January 24, 2012 Cooling limitations are a bad idea, just provides a platform to either lie about results or spend your day accusing someone of cheating. But love the idea of a big structured competition. Well I was thinking in the non extreme side, we both can get LN2 for bench but I also know that is sad enter to a competition and see the top guys running CPUs at high frequencies while you can't even touch the 5GHz (AMD) on air. Just my 2 cents Quote
Massman Posted April 16, 2012 Author Posted April 16, 2012 UPDATE: Start: June 1st. End: September 1st. - 1 competition, 6 sub-competitions, 43 stages. - 19 GPUs, 17 CPUs, 7 MEMs. More exact info soon. Quote
saint19 Posted April 16, 2012 Posted April 16, 2012 UPDATE: Start: June 1st. End: September 1st. - 1 competition, 6 sub-competitions, 43 stages. - 19 GPUs, 17 CPUs, 7 MEMs. More exact info soon. wow, that's big. AMD and Intel? Quote
Luebke Posted April 30, 2012 Posted April 30, 2012 what´s about non-intel/amd/nvidia? 3dfx, s3, sis, cyrix, matrox... Quote
Mr.Scott Posted June 3, 2012 Posted June 3, 2012 The background link is broken. http://hwbot.org/forum/showpost.php?p=177774&postcount=20 Quote
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.