Phantasm66
Posts: 4,909 +8
my post isn't about overclocking. its about no more of this AMD vs INTEL nonsense.....
I have previous posts about what I think about overclocking...
I have previous posts about what I think about overclocking...
AthlonXP2100+ O/C'ing
- Thread starter agissi
- Start date
- Status
sngx_father
Posts: 12 +0
Why not? You are running something out of spec, that is overclocking. There is no difference between a XP2100 and an XP1600 (other than the 2100 was proven to run at that speed), but it's still OCing when you run the 1600 faster. The same goes for this. The core might be the same but it not rated for that speed. Doesn't mean it won't do it though
iss
Posts: 1,994 +1
but that is just the point. they are not out of spec for what the chip is designed for. 9500's and 9700 non pro are UNDERCLOCKED R300 chips that is why the bios flash works.
its the same that ATI did with the 8500 LE's they took R200 chips and underclocked them via the bios. it is a more cost effective method of creating cards priced for different segments ofthe market.
StormBringer
Posts: 2,218 +0
If you want to get technical ISS, Overclocking is changing the clock frequency beyond spec. Any other changes made to get more out of a component is not technically overclocking. The thing is that many of the tweaks overclockers use aren't actually changing clock speed, many other things do. The Term overclocking has pretty much come to include all those things whether directly related to clock speed or not. Changing a device from its "out of the box" condition in order to acheive better performance is considered by most to be overclocking.
The point here is that the card was built and sold with certain specs, by changing them you have changed it from the way it was intended. Just because the 9700 is the same is not relavent because the 9500 was sold at those specs, not at the specs of the 9700. Changing it, even if only by flashing the BIOS, is still running it out of spec.
The point here is that the card was built and sold with certain specs, by changing them you have changed it from the way it was intended. Just because the 9700 is the same is not relavent because the 9500 was sold at those specs, not at the specs of the 9700. Changing it, even if only by flashing the BIOS, is still running it out of spec.
sngx_father
Posts: 12 +0
:grinthumb
thank you SB, saved me the time...
thank you SB, saved me the time...
iss
Posts: 1,994 +1
that is precisely why I was quite specific in saying the spec for the R300 chip not the spec for any card running the R300 chip.
if the R300 chip is designed and manufactured to run at 325 core speed that is it's spec. what ever hardware or bios limitations a particular card maker takes to ARTIFICALLY limit the R300 chip from running at it's designed specs does not change what the spec of the R300 is.
StormBringer
Posts: 2,218 +0
But it does change the spec of the card as a whole. That was the point that was being argued was it not? Changing the specs of the card. You were the only one who was stuck on the chip itself and its specs.
acidosmosis
Posts: 1,310 +0
Ok, so did I make this thread so large, or was it the original poster? lol.. I have a feeling this long thread is my doing, hehe.
iss
Posts: 1,994 +1
My point is that ENABLING a R300 Chip to run at the design spec's it was made to run at by overcoming the DISABLING that a card manufacturer has placed on a particular card for marketing purposes is a different propositon than Overclcoking a 1600 palomino chip to run at the specs of a 2400 Tbred. and it is a valid point whether or not you want to recognize it.
The 9700 spec might be the MAX 9500 spec, but it was not intended to run like this. It could be that ATI felt these chips were inadequate to run at 9700 speeds and then puts them in the 9500 pile.
It would be very silly of ATI to use 9700 grade parts and waste them in a 9500 card. It would be cheaper to manufacturer the 9500 with crappier parts. Perhaps they take use some 9700 parts to use in the 9500 to fill a quota of some sort... If that is the case, then you really don't know what batch your video card is from and it may not even be capable of running like a 9700....
Just hypothetical stuff though.
This is how CPUs work as well. Many times you'll find the cores used for a 1.5GHz and a 1.9GHz are the same ones. The manufacturer just didn't feel that the 1.5GHz cores (although identical to the 1.9GHz!) - with stock cooling - would be stable at this speed. Every CPU core is different, despite being made identically. Some pass off as 1.9GHz, others pass off as 1.4GHz.. etc..
Overclocking takes advantage of this. Older CPUs such as the early Athlon were not "locked", so overclocking them could be done through software. The reason AMD did this - It's the same CPU, only electronically crippled.
Your argument is actually favorable to overclocking, whether you want it or not.
iss
Posts: 1,994 +1
a 1600 palomino is not an underclocked 2400 T bred. a 9500 is an underclocked R300 core. yes manufacturers do use chips that dont perform up to specs as lower clocked products. but it is less common in mature processes like the .15 micron. AND since the 9500 is the mainstream product the 9500's volume alone argues against the bulk of them being poor performing R300 chips.
Justin
Posts: 914 +0
A 1600 palamino is not a tbred, no. It's a palomino. It may have been made on the same piece of silicon as a 2400 palomino, though, and it just so happened that one ran at 2400 stable and the other couldn't.
It isn't less common at all. In fact, it is MORE common then ever before, since the odds of any one produced chip having the exact same rated performance has another chip from the same slab of sillicon are a lot less likely then before. I suggest you do a little more research into CPU fabrication before stating that.
It isn't less common at all. In fact, it is MORE common then ever before, since the odds of any one produced chip having the exact same rated performance has another chip from the same slab of sillicon are a lot less likely then before. I suggest you do a little more research into CPU fabrication before stating that.
StormBringer
Posts: 2,218 +0
What Rick is trying to say, and you ISS, can't seem to understand, is that manufacturers(not just ATI) often use many of the same parts(even CPUs and mainboards) in several products, but they change other components such as resistors and capacitors. They would use cheaper lower rated components in the lesser product because it doesn't produce the same temps or require the same power as the higher end product. This means the are NOT the same product. You aren't going to hear about the passive component differences by reading a review or even an OC tutorial, you'd only see those differences by reading a schematic of the two products, or by comparing them part by part.
iss
Posts: 1,994 +1
uhhh... you are suggesting that a .18 micron chip is made on the same wafer as a .13 micron chip? I think it is YOU that needs to do a bit more study before making statements.
Stormbringer,
the difference in resistors on the 9500/9700 are in ADDITIONAL resistors placed on the board of 9500 to prevent the R300 chip from running at the R300's specs. and either way the additon of different components on the BOARD does not change the Spec for which the R300 was manufactured to run at. in other words the R300 chip doesnt become something other than a R300 chip because of the components on the board.AND rather than not understand this it is precisely what I was referring to when I mentioned cardmakers who artifically limit the R300 chips performance via bios or HARDWARE.
That said, YES differnet components will affect how the R300 chip is able to perform.
StormBringer
Posts: 2,218 +0
You still don't get it! The two cards are not the same. This was the point. No one said that the chip had been changed, they said that the overall specs of the card had been changed. The changing the specs of the card to make the chip run at its full design specs is modifying the card from what it was built to do. It doesn't matter that the chip is capable of doing it. The chip was put on that card to run at the lower spec ON THAT card.
You also fail to understand what I said about the components. I wasn't talking about adding components to drop voltages and modify the output of the chip. I was talking about tolerences and ratings. Just because you need a 2k resistor doesn't mean you can use any 2k resistor. These components have temperature ratings and % tolerence. The tolerence is an accuracy rating of sorts. meaning that the component's actually value may vary by the % specified.
You also fail to understand what I said about the components. I wasn't talking about adding components to drop voltages and modify the output of the chip. I was talking about tolerences and ratings. Just because you need a 2k resistor doesn't mean you can use any 2k resistor. These components have temperature ratings and % tolerence. The tolerence is an accuracy rating of sorts. meaning that the component's actually value may vary by the % specified.
Justin
Posts: 914 +0
No, that was just an example. I don't know the exact micron size of the palamino series, however chips of the same architecture will be produced on the same wafer yet sold at different speeds. By your logic, say you had 2ghz part and a 2.2ghz part off the same wafer, if you upped the speed of the 2ghz part you're not really "overclocking" at all (This is EXACTLY what ATI does with their DSPs and GPUs) - And yet you also say you are shortening the life of the cpu and risking damaging it. If you aren't really overclocking how are you doing that? I fail to see how your logic is put together and it seems very faulty.
Originally posted by Vehementi
You can't change the multiplier at all without unlocking it, which is explained at length why you shouldn't do it in that very article.
I don't overclock usually. 1733MHz is plenty fast for surfing the net and using mIRC
Thats not essentially true.. Older boards that don't know about the newer tbird chips simply ignore the multiplier lock!
As far as the same cores in CPUs go and some being underclocked, it is definitely true.. My 1800+ (1533, tbred core) runs rock stable at 12.5x155 (1937.5) at 1.7v with stock AMD fan.. Even in the hot Australian summer LOL..
Over clocking
i hate it when people say they are to scared to over clock their machienes because they are afraid of what MIGHT happen to their machiene. read all the aritcles through out this web site and others about all the people that have successfully over clocked their computers with no problems at all.
yes i understand that people may not feel comfortable with adujsting SOME settings, there isnt much really if you think about it. if you dont want to over clock that is fine but dont bring down others for doing it.....
i over clocked my AMD 2800+ 2.0ghz to 2.4ghz with NO problems at all, no freeze ups or ANY system errors. there is no harm in over clocking just so long as you know what your doing and its not hard to find out info about doing it. alot of info can be found on this site and all the posh tech sites that do reviews on CPU's all the time..
just think about it..
i hate it when people say they are to scared to over clock their machienes because they are afraid of what MIGHT happen to their machiene. read all the aritcles through out this web site and others about all the people that have successfully over clocked their computers with no problems at all.
yes i understand that people may not feel comfortable with adujsting SOME settings, there isnt much really if you think about it. if you dont want to over clock that is fine but dont bring down others for doing it.....
i over clocked my AMD 2800+ 2.0ghz to 2.4ghz with NO problems at all, no freeze ups or ANY system errors. there is no harm in over clocking just so long as you know what your doing and its not hard to find out info about doing it. alot of info can be found on this site and all the posh tech sites that do reviews on CPU's all the time..
just think about it..
JimShady23
Posts: 360 +4
thats what I was about to say, I was looking through some old trends just for fun and came across this one LoL.
But just to add my $0.02 if you can overclock and keep your temp lower than what a stock cpu with stock HSF will run you will be fine...I was flipping out when my 2500+ OC to 3200+ was running idle at 41c....52c under load....Here I see there are guys running stock AMD's @ 50, 60, even 70 !!!! Idle without problems I felt alot better heh
But just to add my $0.02 if you can overclock and keep your temp lower than what a stock cpu with stock HSF will run you will be fine...I was flipping out when my 2500+ OC to 3200+ was running idle at 41c....52c under load....Here I see there are guys running stock AMD's @ 50, 60, even 70 !!!! Idle without problems I felt alot better heh
- Status
Similar threads
Latest posts
-
Astronauts aboard the ISS craft miles of technically superior optical fiber- Shawn Knight replied