Unofficial Anandtech OC contest discussion: now with poll!

DrMrLordX

Lifer
Apr 27, 2000
21,582
10,785
136
In the interest of not turning this thread into a flame/trollfest instantaneously, I shall respectfully request that discussion of existing events present or future be left out of this thread unless you have been a participant in a past event and want to add your own commentary about things that worked and things that didn't. This thread does not exist to promote or condemn any event that has already taken place, is taking place, or will take place; we just want honest commentary from people who have been to past events either as spectators or contestants (preferably).

What I really want to do is discuss an overclocking event, probably run by the Anandtech community since I have not yet detected any interest on the part of site admins with regard to running an OC contest.

Three major questions loom over the notion of an overclocking event:

1). To whom would the event appeal?

2). How would we appeal to the target audience?

3). How should the event be organized?

#2 and #3 overlap a bit but I think you can tell the difference. Or maybe not.

Anyway, thoughts? I could go on and on but watching myself type shouldn't be the goal here.
 

SickBeast

Lifer
Jul 21, 2000
14,377
19
81
Is this going to be a real event where you have to go there in person? Perhaps you should take a poll to see where most people are from. Maybe you will need to do one in NYC and one in LA.

1. As many people as possible. IMO the best thing to do is to make it the best overclock in terms of a percentage from stock speed, perhaps even at stock voltage.

2. Perhaps offer prizes. :D

3. I guess set up a leaderboard, and let people just show up and show you a Prime stable rig. You could charge an entry fee of $10 to cover the cost of the prizes, similar to a poker tournament.

You could seggregate AMD/intel, or else you could even do it year by year in terms of when the processor was released. If you get enough interest you could do GPU overclocking as well in a similar fashion.

I think you should install Orthos or Prime to a USB memory stick and then use that to test each rig. I guess 15 minutes should be the gold standard seeing as you don't want people to have to wait too long. I guess an hour wouldn't kill anyone; most bad overclocks will fail within a few minutes.

Oh yeah, make it air cooling only, however I would not place restrictions on the HSF whatsoever to encourage ingenuity and creativity.
 

SickBeast

Lifer
Jul 21, 2000
14,377
19
81
An idea might be to have this at one of the big computing conventions like the CES or whatever.
 

DrMrLordX

Lifer
Apr 27, 2000
21,582
10,785
136
Okay, some good ideas from Sickbeast. Anyone else want to weigh in? This is just a brainstorming session at the moment but ideas are fun to toss around at the very least.
 

Shmee

Memory & Storage, Graphics Cards Mod Elite Member
Super Moderator
Sep 13, 2008
7,382
2,415
146
I am not sure that a pure % overclock standard would be fair, as that would just encourage people to find one of those pentium 4 chips that did 100+% oc's with low stock speeds. This defeats the purpose of OCing, which is to gain performance. In stead we might want to have a cpu benchmark with categories, like one for i7, one for c2q, one for dual cores, and one for AMD quads.

I don't see a problem with going above/below stock voltages, after all, if it passes the stability test, it is after all the user's chip. Though we might want to limit it to recommended "safe range" to avoid ppl trying extreme 1-time bench oc's at crazy volts.

Prizes sound good, as do free refreshments :D

Also, limiting it to air cooling might be a good idea, basically making the best cooling you could do would be a lapped TRUE with 2 high speed fans(which I have :DDDD) It would at least keep results in the same general range, so there would be no worrying about sub zero or near zero CPU temps.

Also, we would have to find out the best place to host it. Where do most people live?

...Just some ideas.
 

DrMrLordX

Lifer
Apr 27, 2000
21,582
10,785
136
More good input. I'll post later when I have the time, but feel free to keep the ideas coming in if anyone so desires.
 

imported_Shaq

Senior member
Sep 24, 2004
731
0
0
The problem with this is that it is based on luck and not skill. Whoever was lucky enough to get a good chip would win. Why not have a single CPU and see who can OC it the most? There could be different categories also, but the problem would be purchasing the chips unless it could be borrowed/donated for the competition. The biggest drawback would be time of course. But you could have it set for 1 hour with the chip and make it 20 passes LinX stable or something like that. I think that would be far more interesting. And it could be used with the same heatsink/power supply/RAM etc. that way.
 

Idontcare

Elite Member
Oct 10, 1999
21,118
58
91
Originally posted by: DrMrLordX
This thread does not exist to promote or condemn any event that has already taken place, is taking place, or will take place; we just want honest commentary from people who have been to past events either as spectators or contestants (preferably).

2). How would we appeal to the target audience?

Who are "we"?

Originally posted by: Shaq
The problem with this is that it is based on luck and not skill.

That same argument applies to all forms of competition in every sport.

You don't do Nascar races by having each driver drive the same car in series fashion.

I agree some form of event planning pre-binning/filtering should be done merely to weed out the known/obvious duds (the chips that just take massive Vcore to OC or don't OC much at all).

But restricting the entire field to identical hardware takes some of the unknown out of the equation and some of the unknown is part of the enjoyment of watching the spectacle.

Imagine if every baseball player had to hit with the exact same bat, or wear the exact same glove, etc.
 

DrMrLordX

Lifer
Apr 27, 2000
21,582
10,785
136
Originally posted by: Idontcare


Who are "we"?

The Anandtech community as a whole. Though, I must admit, the first statement from which you highlighted text is more than a bit presumptive on my part.

The idea was to try and steer conversation away from promoting or bashing events that area already taking place or in the planning stages.
 

DrMrLordX

Lifer
Apr 27, 2000
21,582
10,785
136
Okay, several thoughts here based on other replies:

Physical location: An online-only event is tempting but I have no idea how anyone could really control things to the extent that would be necessary. So yeah, some kind of sideshow at an existing trade show or other event like CES might be a good idea.

As far as regions or municipalities are concerned, I could see LA, Las Vegas, Atlanta, Chicago, Houston, or New York as being possible venues; just keep in mind that most of these places have certain costs associated with event space, and that we could get a better deal in a smaller burg. Choice locations could include anyplace that is a major transit hub that is conveniently located with respect to a majority of those among the community that would choose to participate.

That sort of comes down the road though, since we'd have to have some money dangled in front of our noses before we can seriously start picking a venue.

To address the issue of "to whom should the event appeal", while "as many people as possible" is a good goal, we have to admit that large segments of the population have no idea what overclocking is or what can be accomplished via overclocking. I still run into computaphobes that think overclocking might be illegal. Or something.

Overall I'd say our target audience would be site readers, forum readers, and forum posters from communities like Anandtech. There are any number of other hardware enthusiast forums like ours that might be interested (or interested in competing with their own event).

Outside of that, people who don't really "get" computers still sometimes like games. Any overclocking event would have to have a gaming aspect, that is, at least one of the goals would be to get maximum gaming performance out of the rig in the contest. It doesn't mean that gaming would be the "only" focus, but it should be at least part of one event.

As far as "how would the event appeal to the target audience", prizes are good, but we also have to set up the event to interest the target audience before prizes are taken into account. We also have to produce the appearance of legitimacy, preferably by actually being legitimate. Fortunately, several posters here have already touched on that topic.

As far as being legit goes, I would think we'd want all parts used in the competition to be binned from a small selection of commercial samples. Ideally we'd have a vendor like NewEgg, ZipZoomFly, Tankguys, or what have you provide parts from their own copious stockpiles. They could produce an internal process for binning parts, and contestants would pay for the binned hardware as a part of their entry fee. The contestant would take the system home with them regardless of their finish.

CPU, motherboard, and memory would be binned (memory would be binned by kit rather than by DIMM). Vid card might also be binned but I don't want to weigh in on that just yet.

Outside of cpu, mobo, memory, and possibly vid card, PSUs would be stress-tested to make sure we wouldn't have any DOA units to hand out to participants. Otherwise, participants would be able to bring any cooling equipment they wanted as long as it was all air; in other words, they could nail the system to a board and run it passive, or they could bring along a $$$ case, tons of fans, prefabbed ductwork, insane HSFs with massive fin arrays attached, or whatever. Maybe that's the wrong approach, but I'd hate to force people to choose cooling from pre-selected stuff that the vendors were able to bring (cooling stuff can be bulky).

Bottom line is that each contestant would specify a platform, then specify CPU, mobo, and memory (and probably vid card) as a part of their entry, get an entry fee quote from the vendor(s) working with the event coordinators, and then get the binned parts upon arrival. They'd get a PSU assigned to them that had been stress-tested and shown to run within some power envelope either specified by event coordinators or by the entrant (maybe the contestant could pick their PSU too, but stress-testing would still be done to make sure the unit was not DOA), and then they'd have a set amount of time to throw their system together, do lapping, do modding, etc. ala Iron Chef. Anything from dremeling cases to volt mods on components would be allowed. About the only thing we wouldn't let people do is switch out physical components of the board, such as by installing better caps or pulling chips or what have you (though, I don't know, maybe we should let people swap caps if they have the skill?).

I kind of like the idea of including some kind of cost equivalence scale so that the value of outside cooling components could be added to the entire system value for budget and bang-per-buck award categories, but handling that could be sketchy. If it's not too much of a hassle they could source cooling components through the event vendors, but some people are going to want to make custom cases out of wood, lexan, or other materials from a hardware store rather than just carve up a prefabbed case. I suspect that some of the best entries will just be nailed to wooden boards.

Anyway, once built, each rig would be subjected to stability tests administered by event coordinators using the timings and settings supplied by the contestants. Failure would amount to disqualification (unless that's too draconian; timing/speed reductions following event-specified increments might be applied to achieve stability, but this would be done by judges and without oversight from the contestant). Then a pre-defined set of "real-world" benchmarks would be run, and points would be awarded based on where each rig finished in each benchmark.

The total points winner would win "best system", which should be the biggest prize.

The total points winner within a certain budget threshhold would win "best budget overclock". The cost threshhold would either involve the vendor-specified prices of CPU, mobo, memory, and PSU, or all of the above plus other components relevant to the overclock (case, fans, hsf, etc). Again this goes back to a question of how, or if, cooling accessories should be sourced through the same channel. It would stink to see a system with $300 of sourced parts and $2000 worth of cooling win the budget overclock category.

A third category, bang-per-buck, could exist for the system with the best ratio of points to cost.

We might introduce sub-categories for best Intel or AMD system in any of the above categories, and then there's the issue of gaming which could have its own set of benchmarks and a similar award structure.

Whether or not we'd want to include the cost of video card in the primary budget and bang-per-buck categories depends entirely on whether or not the selected benchmark suite would have anything that could be accelerated by a video card. Obviously, video card costs would factor into total system cost for any gaming award categories.

Or, at least, those are my ideas. Thoughts? Also, is there any reasonable way to test for stability that wouldn't involve a 24-hour stability testing event prior to prizes being handed out? That could bore the stew out of people, or just make an excuse for a really wild party leading up to the final benchmarking.
 

imported_Shaq

Senior member
Sep 24, 2004
731
0
0
Originally posted by: Idontcare
Originally posted by: DrMrLordX
This thread does not exist to promote or condemn any event that has already taken place, is taking place, or will take place; we just want honest commentary from people who have been to past events either as spectators or contestants (preferably).

2). How would we appeal to the target audience?

Who are "we"?

Originally posted by: Shaq
The problem with this is that it is based on luck and not skill.

That same argument applies to all forms of competition in every sport.

You don't do Nascar races by having each driver drive the same car in series fashion.

I agree some form of event planning pre-binning/filtering should be done merely to weed out the known/obvious duds (the chips that just take massive Vcore to OC or don't OC much at all).

But restricting the entire field to identical hardware takes some of the unknown out of the equation and some of the unknown is part of the enjoyment of watching the spectacle.

Imagine if every baseball player had to hit with the exact same bat, or wear the exact same glove, etc.

Then all someone would need to do to win is buy 20 D0's, pick the best one and Ebay the rest. I'm not saying it would be worth the hassle to win but that should do it. If there were prize money involved though it might be worth it. OTOH, if everything was standardized you would have to really know your stuff. There are a hundred settings to adjust when including RAM. This is more of a cerebral competition and not like sports. Think of it more like a chess tournament. I don't think they use different boards and chess pieces do they? Otherwise you have a bunch of people speed binning chips for you. Why not have Intel sponser the event and they can rebadge some of the CPU's, resell them and make a tidy profit? lol I'm not being abrasive here, just trying to make a point. I'm all for a competition though I doubt I would go unless you can have it in Memphis, St. Louis or Dallas. Houston isn't too bad though. Although if Aigo goes I'll stay home. lol

OP: LinX can crash unstable settings in under 30 minutes, especially CPU related.
 

Idontcare

Elite Member
Oct 10, 1999
21,118
58
91
Originally posted by: DrMrLordX
Originally posted by: Idontcare


Who are "we"?

The Anandtech community as a whole. Though, I must admit, the first statement from which you highlighted text is more than a bit presumptive on my part.

The idea was to try and steer conversation away from promoting or bashing events that area already taking place or in the planning stages.

Ah, I'm with you now :thumbsup: I don't think its presumptive in the least, but the way it read to me came across as you being a representative of some unmentioned background entity/business/organization so it piqued my curiosity.

So is it safe to say the spirit of the message (in part) goes something like "if we were to engineer an OC event tailored to our interests, what would we like to see in the event and what would we like to know won't be a factor in the event"?

As others have stated, the metric of success in basically all other OC events I have read about ends up being suicide benching which definitely does not appeal to the majority of active members here in the AT CPU subforum.

So yeah definitely this needs to be an OC event where the metrics are clearly stated and include (1) absolute performance, (2) performance/cost, and (3) performance/watt at 24/7 stable settings.

If the rig/setup doesn't pan out as being deemed 24/7 stable capable then the majority of folks here, the "we" as I perceive it, won't really care about the performance or any other metrics of success tied to it.
 

Idontcare

Elite Member
Oct 10, 1999
21,118
58
91
Originally posted by: Shaq
Then all someone would need to do to win is buy 20 D0's, pick the best one and Ebay the rest. I'm not saying it would be worth the hassle to win but that should do it. If there were prize money involved though it might be worth it. OTOH, if everything was standardized you would have to really know your stuff. There are a hundred settings to adjust when including RAM. This is more of a cerebral competition and not like sports. Think of it more like a chess tournament. I don't think they use different boards and chess pieces do they? Otherwise you have a bunch of people speed binning chips for you. Why not have Intel sponser the event and they can rebadge some of the CPU's, resell them and make a tidy profit? lol I'm not being abrasive here, just trying to make a point. I'm all for a competition though I doubt I would go unless you can have it in Memphis, St. Louis or Dallas. Houston isn't too bad though. Although if Aigo goes I'll stay home. lol

Well again I will just go back to the plethora of existing examples in all forms of competition to say regardless of the degree of physical involvement (sports) the cerebral component is there in the form of strategy.

There are two paths to select for an event like this, and any other that involves more than just humans.

Is the event created so as to showcase the hardware or the people?

If the purpose of the event really is to showcase the people then yes you have to stringently control the boundary conditions and initial conditions relating to all other aspects of the event - the hardware, the funding, the access. But guess what happens to your access to money, sponsors, etc when you remove the prospects of making the unstated purpose of the event as a showcasing of the hardware?

Consider the olympics versus the special olympics. One is very much about showcasing the people, the other is really about advertising and politics, the people involved are really just the means to that end.

You aren't going to see raging debates over the technological advantages given to affluent national teams with the speedo LZR Racer in the special olympics, primarily because there is no money to be made by the sponsors there.

So how do you craft an event where the stated purpose, showcase the people and their OC talents, is truly the purpose of the event while maintaining some compelling reason for sponsors (which are not charities) to get involved in the fiscal aspects of the event?

Look at Nascar and how the push to eliminate hardware differences (advantages) has really eliminated much of the differentiating value between the car manufacturers. Now the revenue goes to the owners of labels painted on the hood (Home Depot, RedBull, etc) instead of the companies that engineer what goes under the hood (Ford, Toyota, etc).

OC events have yet to make this transition, the sponsors still very much end up being the same folks who build/sell the hardware itself. As such how do you convincingly eliminate the showcasing of the hardware (the event sponsors) as the unstated purpose of the event while simultaneously convincing the hardware guys to pony up and sponsor the event?

I personally do not believe such a separation can be done cleanly until the revenue model shifts from one of self-promotion (showcasing the hardware used in the event is the point of sponsoring the event) to one of third-party advertising. Meaning you get redbull to sponsor the event, make it a goal to reduce the hardware to merely serving as a means to the end, and showcase the people (as well as the redbull!) as much as possible.
 

cboath

Senior member
Nov 19, 2007
368
0
76
As someone a bit newer to the OC scene - in addition to a competition, a way to appeal to wider audience would be workshops or session on how to OC, what to look for, what to do, what NOT to do, etc. People's descriptions of what they do don't always easily translate across the board, and seeing it first hand (photo's of bios screens are very nice) and the procedure taken can really jumpstart people on the road to success.

As for a competition, i think you'd have to segregate it to brands and versions. Not really fair to have a guy with C2D or C2Q competition against a Phenom or i7, and similar cooling, too. Can't have a guy on water competition against a guy on stock cooler. Apples to Apples as they say. And if this is in a physical location, i think you'd need set hardware to use and not people using their own stuff. The challenge of an online thing is proof. How do you know posted screens are legit?


Location brings up another point, too. If it's a physical location - you can make it a much larger event. You can then recruit the big guns (EVGA, ASUS, Intel, AMD, NVidia, ATi, etc, etc) to have booths and displays, to see if they'll offer prizes etc. You could do that to an extent for an online event as well, i guess.

As noted above, you need reasons for those who aren't the top tier of OCers to show. I suppose you could try different categories - having to rank yourself as beginner, intermediate, advanced, master, etc... But that opens the door for sandbagging your self-rating to have a better chance to win. That's why i'd suggest classes/tutorials on how to OC for people who don't think they can win. Short of that, you could change the format, if physical, to a regional aspect with a traveling core maybe. Aug 1 in Chicago, Aug 8th in Indy, 15th in Pittsburgh, etc. Then Have the top guys from the NE (for example) participate in competition. The NE region winners then go off to the 'nationals' in Dallas (or wherever) to compete against those who were tops in the West, Southeast, and Central regions. Dunno, just a thought. Less travel that way for the majority.
 

DrMrLordX

Lifer
Apr 27, 2000
21,582
10,785
136
Originally posted by: Shaq
Otherwise you have a bunch of people speed binning chips for you.

Binning by the event sponsor(s) was what I had in mind, and is, in my opinion, the most fair way to go about it.

Namely, bin the chips based on their max OC at stock vcore (so as to not stress or damage the chip during binning). The vendor could easily box up the other chips and sell them as new, which they (presumably) do anyway.

OP: LinX can crash unstable settings in under 30 minutes, especially CPU related.

Not a bad idea. I'm still flirting with the idea of a true 24-hour stability test to serve as a front for a wild party or for workshops like one of the other posters in here mentioned (see below).

Originally posted by: Idontcare

Ah, I'm with you now :thumbsup: I don't think its presumptive in the least, but the way it read to me came across as you being a representative of some unmentioned background entity/business/organization so it piqued my curiosity.

I wish I had money/power behind me. Sadly, that is not the case.

So is it safe to say the spirit of the message (in part) goes something like "if we were to engineer an OC event tailored to our interests, what would we like to see in the event and what would we like to know won't be a factor in the event"?

Yessir.

As others have stated, the metric of success in basically all other OC events I have read about ends up being suicide benching which definitely does not appeal to the majority of active members here in the AT CPU subforum.

Bingo.

So yeah definitely this needs to be an OC event where the metrics are clearly stated and include (1) absolute performance, (2) performance/cost, and (3) performance/watt at 24/7 stable settings.

I think 24/7 stability should be established before we bench anything. There should be no suicide bench category.

If the rig/setup doesn't pan out as being deemed 24/7 stable capable then the majority of folks here, the "we" as I perceive it, won't really care about the performance or any other metrics of success tied to it.

I would include myself in that category.

Originally posted by: Idontcare

There are two paths to select for an event like this, and any other that involves more than just humans.

Is the event created so as to showcase the hardware or the people?

The people, I would think, along with their ability to pick the right hardware.

If the purpose of the event really is to showcase the people then yes you have to stringently control the boundary conditions and initial conditions relating to all other aspects of the event - the hardware, the funding, the access. But guess what happens to your access to money, sponsors, etc when you remove the prospects of making the unstated purpose of the event as a showcasing of the hardware?

It depends on who your sponsors are. If you ask a manufacture to sponsor the event (Intel, AMD, Nvidia) then it has to be about the hardware. If you ask a major vendor to sponsor the event (NewEgg, ZipZoomFly), then they're happy as long as they get exposure and improve their rep as a reliable provider of performance parts, regardless of what those parts might be.

That's why I think parts vendors, not manufacturers, should be solicited for financial support. If the event ever got big then the manufacturers would be more likely to sponsor competitors or teams than the event as a whole, just to get in there and try to prove the worth of their own products.

So how do you craft an event where the stated purpose, showcase the people and their OC talents, is truly the purpose of the event while maintaining some compelling reason for sponsors (which are not charities) to get involved in the fiscal aspects of the event?

You source most or all parts for the event through <insertnameofvendorhere> and put up signage everywhere for that vendor, and have event coordinators and competitors (particularly winning competitors) go on and on about how nice it was to source all their parts through such a great vendor. Da nubs looking to do their own OC (and believe me, the number of people that will clone winners will be pretty large if anyone pays attention to the event) will probably buy parts from <insertnameofvendorhere> expecting similar results despite the fact that the contest binning will not be done if they just belly up to the bar and buy some stuff online (plus they will lack the skill of the competitors).

Look at Nascar and how the push to eliminate hardware differences (advantages) has really eliminated much of the differentiating value between the car manufacturers. Now the revenue goes to the owners of labels painted on the hood (Home Depot, RedBull, etc) instead of the companies that engineer what goes under the hood (Ford, Toyota, etc).

Exactly. And look at who gets their name on events as primary sponsor - it's Nextel, not Ford or GM.

OC events have yet to make this transition, the sponsors still very much end up being the same folks who build/sell the hardware itself. As such how do you convincingly eliminate the showcasing of the hardware (the event sponsors) as the unstated purpose of the event while simultaneously convincing the hardware guys to pony up and sponsor the event?

Simple: don't ask the hardware guys to pony up the cash for the event. Let the middlemen do it. There are some big middlemen out there, and they stand to reap great rewards from promoting overclocking so long as they can credibly sell themselves as a major supplier of OC-friendly parts.

I personally do not believe such a separation can be done cleanly until the revenue model shifts from one of self-promotion (showcasing the hardware used in the event is the point of sponsoring the event) to one of third-party advertising. Meaning you get redbull to sponsor the event, make it a goal to reduce the hardware to merely serving as a means to the end, and showcase the people (as well as the redbull!) as much as possible.

That is another possibility. If you make it a gamer-oriented event, or partly gamer-oriented, you could get sponsorship from any number of gamer-gear companies that supply stuff with no real relevance to overclocking (Bawlz, Mountain Dew, Red Bull, ThinkGeek, etc.)

Originally posted by: cboath
As someone a bit newer to the OC scene - in addition to a competition, a way to appeal to wider audience would be workshops or session on how to OC, what to look for, what to do, what NOT to do, etc. People's descriptions of what they do don't always easily translate across the board, and seeing it first hand (photo's of bios screens are very nice) and the procedure taken can really jumpstart people on the road to success.

I think that's a great idea. Workshops could be carried out during the 24-hour stress test segment, if people aren't all busy getting drunk or soliciting hookers (you laugh, but if you've ever heard stories from E3 in years past . . .)

As for a competition, i think you'd have to segregate it to brands and versions. Not really fair to have a guy with C2D or C2Q competition against a Phenom or i7, and similar cooling, too. Can't have a guy on water competition against a guy on stock cooler. Apples to Apples as they say.

While I agree on the cooling aspect (which is why I want it all-air, and I want to give competitors some leeway with what they choose to use while rating their choices on some kind of a cost scale), I don't agree on the platform aspect, per se. If you include budget and bang-per-buck categories, I think people can compete with any platform they choose, even if you don't segregate based on manufacturer. Obviously the maximum performance overall will come from an Intel rig, right now, but who would win the budget category? I'd have to say AMD users will win that. Bang-per-buck? That's anybody's guess.

And if this is in a physical location, i think you'd need set hardware to use and not people using their own stuff. The challenge of an online thing is proof. How do you know posted screens are legit?

Yeah, online would be hard to do. I think we need to source all hardware through the event sponsors (which is why I'd like a vendor like NewEgg to sponsor it).

Location brings up another point, too. If it's a physical location - you can make it a much larger event. You can then recruit the big guns (EVGA, ASUS, Intel, AMD, NVidia, ATi, etc, etc) to have booths and displays, to see if they'll offer prizes etc. You could do that to an extent for an online event as well, i guess.

That might come down the road, but for the reasons idontcare articulated, it's probably for the best that actual manufacturers not have an interest in the event, at least not at the get-go. However, running this kind of thing at a larger show would be a good idea if only to increase exposure.

As noted above, you need reasons for those who aren't the top tier of OCers to show. I suppose you could try different categories - having to rank yourself as beginner, intermediate, advanced, master, etc... But that opens the door for sandbagging your self-rating to have a better chance to win. That's why i'd suggest classes/tutorials on how to OC for people who don't think they can win. Short of that, you could change the format, if physical, to a regional aspect with a traveling core maybe. Aug 1 in Chicago, Aug 8th in Indy, 15th in Pittsburgh, etc. Then Have the top guys from the NE (for example) participate in competition. The NE region winners then go off to the 'nationals' in Dallas (or wherever) to compete against those who were tops in the West, Southeast, and Central regions. Dunno, just a thought. Less travel that way for the majority.

Again, that might come down the road. I do like the idea of having tutorials at the event to appeal to those who do not feel like they can win the competition or just don't know what overclocking is.

Having an entire series of events would be nice but you have to have one before you can have many.
 

imported_Shaq

Senior member
Sep 24, 2004
731
0
0
Originally posted by: Idontcare
Originally posted by: Shaq
Then all someone would need to do to win is buy 20 D0's, pick the best one and Ebay the rest. I'm not saying it would be worth the hassle to win but that should do it. If there were prize money involved though it might be worth it. OTOH, if everything was standardized you would have to really know your stuff. There are a hundred settings to adjust when including RAM. This is more of a cerebral competition and not like sports. Think of it more like a chess tournament. I don't think they use different boards and chess pieces do they? Otherwise you have a bunch of people speed binning chips for you. Why not have Intel sponser the event and they can rebadge some of the CPU's, resell them and make a tidy profit? lol I'm not being abrasive here, just trying to make a point. I'm all for a competition though I doubt I would go unless you can have it in Memphis, St. Louis or Dallas. Houston isn't too bad though. Although if Aigo goes I'll stay home. lol

Well again I will just go back to the plethora of existing examples in all forms of competition to say regardless of the degree of physical involvement (sports) the cerebral component is there in the form of strategy.

There are two paths to select for an event like this, and any other that involves more than just humans.

Is the event created so as to showcase the hardware or the people?

If the purpose of the event really is to showcase the people then yes you have to stringently control the boundary conditions and initial conditions relating to all other aspects of the event - the hardware, the funding, the access. But guess what happens to your access to money, sponsors, etc when you remove the prospects of making the unstated purpose of the event as a showcasing of the hardware?

Consider the olympics versus the special olympics. One is very much about showcasing the people, the other is really about advertising and politics, the people involved are really just the means to that end.

You aren't going to see raging debates over the technological advantages given to affluent national teams with the speedo LZR Racer in the special olympics, primarily because there is no money to be made by the sponsors there.

So how do you craft an event where the stated purpose, showcase the people and their OC talents, is truly the purpose of the event while maintaining some compelling reason for sponsors (which are not charities) to get involved in the fiscal aspects of the event?

Look at Nascar and how the push to eliminate hardware differences (advantages) has really eliminated much of the differentiating value between the car manufacturers. Now the revenue goes to the owners of labels painted on the hood (Home Depot, RedBull, etc) instead of the companies that engineer what goes under the hood (Ford, Toyota, etc).

OC events have yet to make this transition, the sponsors still very much end up being the same folks who build/sell the hardware itself. As such how do you convincingly eliminate the showcasing of the hardware (the event sponsors) as the unstated purpose of the event while simultaneously convincing the hardware guys to pony up and sponsor the event?

I personally do not believe such a separation can be done cleanly until the revenue model shifts from one of self-promotion (showcasing the hardware used in the event is the point of sponsoring the event) to one of third-party advertising. Meaning you get redbull to sponsor the event, make it a goal to reduce the hardware to merely serving as a means to the end, and showcase the people (as well as the redbull!) as much as possible.

But the hardware will still be there. You can have banners, sponsors and tables to sell hardware and answer questions. But remember that this is an overclocking event. Most hardware manufacturers don't honor warranties for this nor would offer any technical expertise. So how could Intel or AMD sponsor an event like this? RAM and motherboard manufacturers are a little better in that regard however and they might be willing to sponsor these events.

And you could arrange it so different categories had different sponsors but using the same hardware for each category. That way all the hardware sponsors would win also as they won't be directly competing with each other in the intra-category competition, only overall categories like peformance per watt etc. I thought this would be more of a friendly competition using primarily air cooling that would appeal more to the masses and not the extreme overclocking types. I believe that would work the best and appeal to the most people and the vendors. It would need to be a learning experience if the general public were invited.
 

DrMrLordX

Lifer
Apr 27, 2000
21,582
10,785
136
Anyone else want to comment? One of the things that would determine whether or not an event of this nature would be viable would be interest levels, and at the moment, I am not so sure of the level of community interest in such an event.
 

Idontcare

Elite Member
Oct 10, 1999
21,118
58
91
Originally posted by: DrMrLordX
Anyone else want to comment? One of the things that would determine whether or not an event of this nature would be viable would be interest levels, and at the moment, I am not so sure of the level of community interest in such an event.

You need a poll to generate some feedback regarding potential participant pool-size.

How many people will (a) have the time to attend or participate in the event, (b) have the resources to pay hotel/food/rental car + misc, and (c) if participating then how many have the budget to buy the hardware at the event?

(if you do go the poll route, I recommend adding a couple counter-vote options to give you info too - as in (1) I would like to participate but the travel/lodging expenses will be too much, or (2) I would like to participate but I do not need another computer so the requirement to buy the hardware for the event at the event is a show-stopper for me, etc)
 

DrMrLordX

Lifer
Apr 27, 2000
21,582
10,785
136
Hmm, the response has been less-than-overwhelming, but at least now we know whether or not this topic merits much discussion. Thanks to those of you who chose to weigh in and even to those of you who did not.

At least based on the responses, it seems that the idea sounds good but there isn't much enthusiasm for participating or attending, at least not at this time.
 

aigomorla

CPU, Cases&Cooling Mod PC Gaming Mod Elite Member
Super Moderator
Sep 28, 2005
20,839
3,174
126
problem is this..

who are the highest ocers on AT? well, i think im including in that, and i can think of a few others... which brings you to problem number 2.

Which team are we more associated with? Well, when it comes to xtreme overclocks, and xtreme cooling, i am fairly blasted on XS Forums. :p

I think the same can also be said on the other guys who do have uber high overclocks, they represent another more xtreme geared forum.

I said this before, this is not your highly technical, if you dont have 4ghz+ your not welcomed, Forum. This forum is a nice welcome mat to the general audience who want to get started
in this hobby we call overclocking and tweaking.

From here you get annexed to the more xtreme forums.

I show off new cpu's and new gear on this forum on a entry level. (modest vcore bump, ramping multi, keeping a moderate to mid level QPI.)

I bust out my chiller, and overclock like madness and make people go OMG on XS. :p
 

DrMrLordX

Lifer
Apr 27, 2000
21,582
10,785
136
Yeah, you do have a point there. I guess the reason why I personally was interested in seeing the Anandtech OC community discuss an OC contest (or possibly even run one, or at least comprise a majority of entrants and/or spectators) is that a great many people here, in my opinion, represent what I consider to be the more pragmatic element of the OC community online who could, by virtue of their own personal way of approaching the entire idea of overclocking, vastly improve the OC competition scene.

As it stands, OC competitions are geared to showcase overclocking at a level that many of us can not afford and which often represents suicide-runs intended to show off hardware rather than the individual.

Interestingly enough, the pragmatism inherent to the overclocking ethic here sort of goes hand-in-hand with not having the time, money, or interest in attending an event such as this, so these results are to be expected.

Leaving OC events such as this to communities like XS will be a good thing, I think, in so far as getting them away from being biased showcases for specific manufacturers, but the level of overclocking employed and the associated costs will take the events far out of reach from your average overclocker. Personally I find it hard to get all that interested in seeing an event where competitors employ phase change or LN2 or what have you, since I can not afford to deploy such technology nor can I sustain the power costs associated with such cooling methods.
 

Idontcare

Elite Member
Oct 10, 1999
21,118
58
91
We certainly represent a larger $-volume market segment composed of less-than-xtreme OC'ers, that's for sure. Whether that appeals to potential sponsors or not is another thing.