• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Benching Contest Ideas and Volunteers wanted.

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
Teams would be nice but I doubt we would make more than 3-4 teams and it still won't guarantee that all team members will be active.

There have to be 2 categories for Intel and AMD or it will never be fair. Going back to my i3 example, haswell i3 on stock is beating 4 thread FX/APU @4.5GHz in most tests so we can't simply make clock limits. On the other hand we can't leave it without clock limits as air/water benchers won't have any chance with all who use sub 0.

We could make a poll like what hardware you wish to bench most or which generation of hardware you could use for competition and then pick most popular series divided on categories.

I think we could ask software manufacturers to sponsor competition based on their benchmarks like reward in Futuremark benchmark could be 3DMark or PCMark key or in memory test could be AIDA64 key etc. I bet someone will help us especially that reviewers have some contact with benchmark developers.
The same with hardware. Maybe not anything really expensive but I bet that there could be a chance on coolers or mice/keyboards.

Shipping is other issue as not all will send everything overseas, that's why I think that software keys is not that bad idea and still gives some additional motivation.
 
I have not benched AMD in forever, no idea what they are all about any more, are they competitive clock for clock in 3D ? or do they get hammer there as well ?
 
AMD is worse in nearly everything. In some 3D tests they're even not that bad but Intel still wins. In most other tests it's like I said:
Intel 2 core + HT = AMD 4 threads +1GHz or Intel 4 core = AMD 8 thread +1GHz

I was comparing performance recently as I had i3 4330 3.5GHz and A8 6600K @4.5-4.8GHz.
There is also no space to compare AMD to 6 core+ Intels. AMD gets destroyed by overclocked IB-E in every multithreaded test including 3DMark physics tests. In tests like Vantage, CPU results are maybe 3x higher on 6 core IB-E than on any available AMD.
 
Last edited:
AMD is worse in nearly everything. In some 3D tests they're even not that bad but Intel still wins. In most other tests it's like I said:
Intel 2 core + HT = AMD 4 threads +1GHz or Intel 4 core = AMD 8 thread +1GHz

I was comparing performance recently as I had i3 4330 3.5GHz and A8 6600K @4.5-4.8GHz.
There is also no space to compare AMD to 6 core+ Intels. AMD gets destroyed by overclocked IB-E in every multithreaded test including 3DMark physics tests. In tests like Vantage, CPU results are maybe 3x higher on 6 core IB-E than on any available AMD.

Man that blows, I knew they were behind the curve, but thought they were closer than that

300% deficit would be too much to bare in a head to head. Wish they would put out something completive, this price performance argument makes me want to puke. It's like they gave up.

O-well back too it. I really am out of ideas on how we could make an even playing field based on this, for 3D your statement basically says if you would run an Intel E at 3 GHz the AMD rigs would have to run at 9GHz to compete in the CPU portions of the modern 3D benchmarks :shrug:

So AMD and Intel have to be separated be it 2d or 3D

:bang head
 
Yup. Its got to be classed as AMD and Intel outside of Unigine Heaven really... AMD gets kicked in the gut in 06/Vantage/11, and 3DM (Fire Strike) and of course the 2D's...

We had it this way prior. IMO, that should not change. About the only thing that should change is to limit cooling to ambient so everyone has a fair shot. Its tough to 'force' people to buy a card, even if it is a $30 purchase. That is why I like making a 3/4 person team with someone that has AMD stuff and someone that has Intel. That should also help the 'I have a weak GPU' scenario as well. We can all list our hardware and see if we can group teams fairly with that?

As far as 3D... there is nothing one can do there really. Heaven/FS is an AMD benchmark because you can disable tessellation, otherwise they are on an even playing field.
 
Looks like Catzilla is running better on AMD CPUs than Futuremark benchmarks but still Intel is faster. Also in Catzilla results on Nvidia GTX780/780Ti are much better than on AMD R9 290/290X so I bet that the same will be with lower series.
 
Looks like Catzilla is running better on AMD CPUs than Futuremark benchmarks but still Intel is faster. Also in Catzilla results on Nvidia GTX780/780Ti are much better than on AMD R9 290/290X so I bet that the same will be with lower series.

You're right Woomack, AMD CPU does run better with Catzilla than any other 3D bench I've tried so far But I still had to crank the FX up to 5.4 to edge out my 4770k @ 4.8. In the 1440p test the FX did much better. At the same speed it topped the i7 score by 20%+
 
OK, I am coming up with some ideas.

Perhaps a contest with two man teams, one AMD and one Intel, one nvidia one radeon, one extreme cooling and one ambient on each team. A couple 3d's and a couple CPU benches.

I will work on which benches, format and a scoring method. Maybe a couple weeks per benchmark. Anyone who wants to help organize and keep score/ plan/ moderate, volunteer!
 
I like that idea... but scaled back a bit...

Not sure we have enough people that can go extreme to have teams...the more variables used, the less 'matches' you have for a team.

That said..... I really need to dig out my PM from earlier to JOhan, LOL.....Just have two man teams. The requirements for the team are simply:

*One AMD CPU, one Intel CPU.
*One AMD GPU, and one NVIDIA GPU.
*Ambient cooling only.
*Single card and single GPU on a card only.
*Each person competes in their 'class' (Classes are divided by AMD and Intel CPU only)
s). The "team" score is an average of the two individual scores. For example, if we assign a point value to 'places', say 1 for First, 2 for Second, etc... and Teammate A came in second in AMD while teammate B came in first in Intel, that is a 1.5 average.

What benches, no clue, but, for some reason, I like the sound of that. :p
 
At classicplatforms.com, I made this months comp like this.

This month, First one for this year I'd like to invite enthusiasts to a friendly stand off between hardwares.

s939 - sAM2 - sAM3+

LIMITATIONS!

Single core running operating system only. That means shut cores off on x2 and higher core count processors k?

NO PHENOM II ALLOWED on either AM2 or AM3 platforms. FX (AM3) is however invited.

90nm 65nm and 32nm ONLY -- NO 45nm allowed athlon sempron or other.

Clock speed and cooling is limited to liquid cooling UNCHILLED meaning ambient temps are a must. You can however alter your ambient temps!

Only one submission needed the same way you would at HWBot.

Fasted 32M time on any of these three platforms wins.

You may bench on all 3 platforms, fastest of the three counts only.
 
Why not have handicaps for older equipment, running on air, anything other than an i7 or Titan.

Or catagorize by Brand and cooling.

EDIT: I like Woomack and S_B's ideas too.
 
How, exactly, would you handicap for 'older equipment'?

I like SB's idea as well but like buying new cheap hardware for a small comp, a bunch of people would need said hardware in hand to compete or again be forced to buy it to compete.

IMO, this really should be a 'run what you brung' type competition. Any widdling down should be in the light of having the ability to get MORE people in (for example, just straight up limiting extreme cooling).
 
It looks like team cup is about to start. Everyone is invited to participate in that. It looks like a good variety of benchmarks and hardware for that. Including AMD, intel , cell phones and raspberry pi. Old stuff and new stuff.

I am not going to start any contest until that is done. Everyone is welcome to personally challenge others or also submit to the low clock 32m thread too.
Hwbot thread:
http://hwbot.org/news/10899_hwbot_team_cup_2014_announced_30_stages_from_june_1st_until_august_31st/
Team thread:
http://www.overclockers.com/forums/showthread.php?p=7686281
 
How about a $10-$30 GPU? HD4350 Ebay Link.

Cheap to buy. Can do three or four rounds on the damn thing, and best of all with that slow a GPU CPU choice doesn't matter nearly as much.
See the HD3450 contest for example/proof on that one. I won it with a PhII 955 on air.
 
How, exactly, would you handicap for 'older equipment'?

I like SB's idea as well but like buying new cheap hardware for a small comp, a bunch of people would need said hardware in hand to compete or again be forced to buy it to compete.

IMO, this really should be a 'run what you brung' type competition. Any widdling down should be in the light of having the ability to get MORE people in (for example, just straight up limiting extreme cooling).

Probably the best way to handicap for this senario is to use a set field and use 90% of the difference. For instance lets say you decide to enter with an AMD Athlon64. If the benchmark for this week is SuperPi and the catagory you're competing in is AMD, you can look at HWBot (the field) and compare the top AMD entry to the top Athlon64. You would then take 90% of the difference and add that to your score.

The main downside to using a hadicap system is the additional work for the administrator. However, it will make a much more level playing field.

Also, it hadicaps not only older equipment, but FX might get a handi from Phenoms in some catagories. I can't say the same would be true for intel...I don't think they've ever gone backwards in performance like AMD has. LOL
 
That'd be doable, but a complete pain in the rear for whoever is running it.
Not only does every benchmark have to be handicapped for every chip, but you have to figure out what the top AMD score is, the top Athlon64 score, the top Athlon64 chip, etc. If you're doing it on a per-chip basis the trick becomes to find an obscure chip that nobody has OC'd much. Then you can come in with a score above the best and have a positive handicap to a chip that was run on LN2!
If you only use air results you run into people who selected "Air" or "Stock" when they were running water/etc.
It's a nightmare.
 
Well, looks like this will be on hold for a while.
This thread has been politically correct thus far,

But

How about a knock down drag um out, your hardware will not survive battle to the death contest :chair:

:)

Just throwing this out there :shrug:
 
Isn't that what they all are?

This is my winning 3450 from that contest. Paid $11 for the card I think.

hd3450.jpg


That contest was more fun than the vast majority of bigger, high buck hardware, contests (and benching) that I've done. I think my entire destroyable rig (PSU? Not going to die. Heatsink? Nope.) for that was around $300.

Cheap hardware lets anybody pretend to be a big dog and bench hardware till it dies :D
 
Back