Ideal use for Overclocking - testmy.net resource / tool

resi3js

TMN Friend

Offline

Gender: Male

Posts: 311

Ideal use for Overclocking

« on: November 12, 2007, 01:59:57 PM »

So, here's the situation...

I have an EVGA Nvidia Geforce 8600 GT

In the same line is an EVGA Nvidia Geforce 8600 GT SSC

As you can see, physically they are in no way different from each other. Same fan, same size, probably the same photo. The difference being is that the standard GT uses the stock speeds, while the SSC is preoverclocked. Wouldn't it make sense to buy the standard card, then overclock it to where the SSC is at? I mean EVGA did it, why can't I just to save a few bucks? cheesy

The possible limiting factors I can think of, but cannot see are the voltages (Doubt they're different as that's based on the PCI-E slot, right?) It seems totally possible, doesn't it?

Another thing is that I have an AMD Athlon 64 X2 4000+. Wouldn't it be possible to overclock that to the same speed as the AMD Athlon 64 X2 5000+?

In overclocking the CPU I definently can see limiting factors, the voltages are the same, but here we have multipliers and a FSB which effects the HT and the memory clock speeds, but of course my memory is corsair with heatsinks so it'll probably handle that pretty well, I can probably manually configure the memory to the EPP standards too!

So, what do you guys think?

Btw, my average temperatures are: Graphics card: 50 C (Full load: 60 C), CPU: 18 C (Full Load: 26 C [That's crazy low, don't you think? Even while Cool'n'Quiet is off too!

]) and System: 30 C always.


	Logged

resopalrabotnick

News Anchor
TMN Sr. Veteran

Offline

Gender: Male

Posts: 4501

Troll by nature

Re: Ideal use for Overclocking

« Reply #1 on: November 12, 2007, 02:09:31 PM »

the reason that that will not always work (it could, just saying it does not have to) is that likely the preoverclocked card uses components that were checked out to be able to stand the higher speed by the manufacturer. the reason the "stock" speed is lower than the preoverclocked speed is that the preoverclocking takes a sampling of chips for which the overclocked speed is the equivalent of the "stock" speed for the larger group of chips that are sold running at "stock" speed. since the manufacturer wants to have the biggest yield possible from production he picks a speed, defines that as stock and has a lot of chips that can run it. some of them may be running at the ragged edge just short of frying, for some it may be just right and some could do more. the ones that could do more are the ones that are well and truly overclockable. this is also why some cards will not let you overclock at all without crashing. if the chips on them are already running at their max speed because they happen to be at the low end of the quality curve then overclocking them will put them over the edge.


	Logged

click me! me too!
Give a man a fire and he is warm for a night, but set fire to him and he is warm for the rest of his life.

sig done by S7R1D3R

resi3js

TMN Friend

Offline

Gender: Male

Posts: 311

Re: Ideal use for Overclocking

« Reply #2 on: November 12, 2007, 02:31:17 PM »

I said that the SSC and the standard EVGA cards are physically identical, 100%. same chips and everything, just one has a preOC'd bios,

btw, how many people get a Hypertransport sync flood error occasionally? This is my second one in 2 weeks of having this PC. Confused


	Logged

resopalrabotnick

News Anchor
TMN Sr. Veteran

Offline

Gender: Male

Posts: 4501

Troll by nature

Re: Ideal use for Overclocking

« Reply #3 on: November 12, 2007, 02:44:32 PM »

they have the same type of chip. but a given production run produces a certain amount of dud chips. these dud chips are of course discarded (usually for recycling). the chips that do work are not all identical. flaws that are minor enough to not turn it into a dud can still cause it to perform sub-par. to be able to use these sub-par chips the manufacturer sets a "stock" speed that will allow the greatest number of chips to be "harvested" while reaching the performance goals set.
to build preoverclocked cards the chips that are as close to perfect as possible are selected and put together on cards that can then be run at higher speeds.

so while the chips may carry the same designator and be built from the same design the unavoidable variance during a production run makes them quite different.

same goes for cpu's and the like. if intel makes a 2 GHz cpu only a fraction of the run will operate at that speed without problems. those that don't are then labeled as slower clock speeds according to how fast they can go and sold at a lower price than the top chips thereby still making some money.


	Logged

click me! me too!
Give a man a fire and he is warm for a night, but set fire to him and he is warm for the rest of his life.

sig done by S7R1D3R

disturbed

News Anchor
TMN Veteran

Offline

Posts: 2095

I am and always will be disturbed.

Re: Ideal use for Overclocking

« Reply #4 on: November 12, 2007, 07:54:52 PM »

the thing is that what happens is this:....

intel for example produces a batch of processors that are meant for lets say 3.0GHz - tests them.......some of them might now quite make 3.0GHz - lets say they prove to be working at 2.95GHz.......what are they to do ? throw out that batch that didnt make it ?....no....they simply stamp them as 2.8GHz and sell them for $50 bux cheaper....same thing happens with video cards, etc...

.....its in our advantage to buy those and exploit the way industry functions


	Logged

Are you down with the sickness !?

[ http://hackaday.com ]

disturbed

News Anchor
TMN Veteran

Offline

Posts: 2095

I am and always will be disturbed.

Re: Ideal use for Overclocking

« Reply #5 on: November 12, 2007, 07:56:04 PM »

Quote from: resopalrabotnick on November 12, 2007, 02:44:32 PM

....i guess i didnt see the reply above - basically the same thing I mean


	Logged

Are you down with the sickness !?

[ http://hackaday.com ]