AthlonXP2100+ O/C'ing

By ---agissi---
Jan 18, 2003
Topic Status:
Not open for further replies.
  1. Justin

    Justin Newcomer, in training Posts: 1,595


    Not true.

    If a cpu is using the same die size, same architecture, and same materials, the one that is already rated to run faster is going to run hotter, period. By bringing a chip up to equivalent speeds, with just a little extra money for cooling, you may or may not exceed that temperature. If the cpu is only a few degrees hotter, it won't make that big of a difference. And even if it was a good 10, maybe 15 degrees hotter, it still is a moot point - The life of the CPU is most likely still far beyond the amount of time you would own your system.
  2. RustyZip

    RustyZip TechSpot Paladin Posts: 415

    Can i butt in...??:)

    I myself don't think overclocking is worth it....

    Unless you get to the stage where you have the fastest CPU & Memory that the Motherboard can take... Then instead of spending major wonga on a new Motherboard, CPU and Memory you can overclock - Then Overclock...

    Other than that, i can't see that it's worth it...

    Just my shillings worth...
  3. Rick

    Rick TechSpot Staff Posts: 6,304   +52 Staff Member

    I get frequent comments about how silent my system is - I too enjoy noise much lower than the stock heatsink fan - And I do NOT have a stock heatsink/fan.

    I have such a nice heatsink, I'm actually able to use a slower, quieter fan than what's on the retail HSFs! :) And cooling to spare...



    Not true. Water cooling is as cheap as 50 dollars these days... And there is often no need for water cooling. Fan cooling is more than enough to save money when compared to an equivocally clocked retail CPU.

    As for the warranty, I again would like to mention the MTBF is about 100,000 on modern processors. They are designed to last decades and will do so even when overclocked.

    I have an outrageously overclocked P-III system that runs nearly 24x7.. It has been for 3 years and without any problems. I know other people with simliar scenarios. I personally do not know anyone who has "fried" their CPU prematurely because of overclocking.
  4. Justin

    Justin Newcomer, in training Posts: 1,595

    Typically, a processor that dies from overclocking, at least in my experience, will die within the first week, rather then a few months down the road. As long as the temp is reasonable, if its survived a week it will likely survive many years.
  5. Cucumber

    Cucumber Newcomer, in training Posts: 203

    What the hell do you do with them? I have had my 1.33 running at 1.5, for 1 1/2 years, with slightly raised Voltage! No problem at all with it:) especially since its running on a heatsink which cost me $25 from Frys
  6. Justin

    Justin Newcomer, in training Posts: 1,595

    Well, I have overclocked many many many cpus. I should consider myself lucky for how many that have survived. A few times I have done things a little extreme, and the cpu didn't survive.
  7. Phantasm66

    Phantasm66 Newcomer, in training Posts: 6,504

    I think that different people have different reasons for doing different things.

    If you have what you see as legitimate reasons for overclocking (extend life of system, for fun, as an experiment, to better understand the hardware, to bragg, etc) than that's fine because its your machine and yours to configure in any way that you please.

    I NEVER overclock. Its not because I think its a bad idea, or that when done properly it can't be effective. It can be and it can be. Its just not something I have ever had an inclination to do. Its not something that I am interested in. I am also not particularly keen on adding another stability problem to the already bubbling melting pot of software bugs, etc. I run my machine 24/7 and even although its entirely possible to overclock and still achieve stability, I don't really think I trust myself or know enough about proper cooling or even have the time to investigate.

    Its just a personal choice. I respect people who have done the sufficient research and work to effectively overclock in a stable manner, its just not something that I personally choose to do.

    This does not then, however, make overclocking a stupid idea for other people. It isn't. They have different ideas, expectations and skills and for them perhaps it is viable and sensible.

    Your computer is there for you to investigate and explore. We all take different approaches to that. Whichever you choose is up to you and its stupid to condemn others for not agreeing with your particular viewpoint.

    Just my 10 cents...

    Hope it made sense...
  8. ---agissi---

    ---agissi--- TechSpot Paladin Topic Starter Posts: 2,382   +15

    Yeah it made sence...:D I wont be overclocking, just because I dont thinks its worth my time and the $70 its gonna cost me. Plus I just bought a 2100 for $100(thats /w a fan)....If I had forked over $70 right then and there I coulda got the 2400 at 2Ghz. So I could of got a 2Ghz CPU for the same price as a 1.7Ghz CPU and a BiG cooling solution. I doubt I could o/c past three or 400Mhz anyways. So it would end up taking me more time o/c'ing ...this was in inresting debate, but I think we've got all sides of it now.
  9. Justin

    Justin Newcomer, in training Posts: 1,595

    Overclocking definately is not for the inexperienced PC user. Unless you know an extensive amount (And i mean actually know, not pretend you know, not think you know) of knowledge about hardware in general and an extensive amount about your specific hardware, you should not be overclocking. Also, if the system isn't yours entirely, you also should not overclock it - Don't risk another persons equipment.

    And if you are entirely at all interested in that warranty of yours, you should of course not overclock. IMO it is immoral and unethetical to damage a product due to misuse then return it saying it died for no reason.

    Overclocking isn't for everybody... however it irritates me to no end when people assume it is pointless.
  10. StormBringer

    StormBringer Newcomer, in training Posts: 2,871

    If I may add to that Soul, not only should you "know" how it works but you should "understand" how it works. Just because someone can quote facts and figures, or list the steps in doing it, doesn't mean they understand how it works or could even do it in actuality. There is a big difference in reading about how to do something and actually doing it. Often times you find that your particular situation is a little different than what you read about and you have to adjust the procedure to fit.
  11. Phantasm66

    Phantasm66 Newcomer, in training Posts: 6,504

    The main reason I don't overclock, I think is this....

    Here is a little story on what happened to me the other day.

    I was happily downloading some files. I was sitting on some xdcc queues on IRC for some files. On some of the queues, I had been waiting for over 1 day.

    I placed an EXTREMELY cheap CD-R into my Writer. And...

    The machine COMPLETELY LOCKED UP trying to mount the crappy, cheap CD. No mouse movement was possible. CTRL+ALT+DEL did nothing. NUM LOCK key would not toggle NUM LOCK light on or off.....

    :evil: DEAD SYSTEM :evil: !!!!

    My point is that there are enough problems with PC hardware and software right now, without introducing any more, particularly when that's going against the manufacturer's stated recommendations.

    BUT THAT'S JUST ME AND HOW I LIKE TO OPERATE.

    Plenty of stories of people overclocking with no problems at all, got themselves a good few extra MHz for next to nothing and learned something in the process. Its just that I don't think, and never have thought, for some reason that I would be one of them.

    I've made the choice not to, but to dub the whole thing POINTLESS would be stupid, because as we've already discussed, often it isn't.

    Mind you, should probably not be using cheap CD-Rs as well ;) !

    Hope this post made sense as well...



    Indeed. There's a world of difference.

    Right now, I am studing computer network administration, lots of exams and test on how we would set up WAN, etc...

    All is fine on paper, but when you come to try it, its another story! Just because you can pass an exam on something does not mean that you can actually DO IT.

    Recently I've come to realise that I've rather naively been underestimating just how hard it is to do certain computing tasks.

    If the money to buy another mobo and chip is a lot to you, then don't mess with it unless you know EXACTLY what you are doing....
  12. sngx_father

    sngx_father Newcomer, in training Posts: 17

    ISS i think you're missing the point. I grant there are people out there who live and die by their 3D Mark 2k1 score, but the maiin motivation of OC'ing is to get the most out of what you pay for. It all started back when people realized that there was no difference between a P100 and a P133, other than that the P133 was proven to run that fast.

    OC'ing has always been about seeing how fast you could make your system run SAFELY. There are *****s out there of course who use LN2 to OC 200% for 5 minutes... just long enough to run some benchmarks. Even then they're just exploring the limits of the technology.

    If you read techsites better, you will see there are mixes of real-world app benches (3D Games, encoding) and artifical benches (Sandra, etc.). That is because the testers know that artificial benchmarks are not the be all end all. Hell even Sandra says that in their documentation...

    Now saying that enocoding is not something most people do, many people rip CD to MP3 everyday, and that is an app that is completely CPU bound. OCing the CPU helps greatly in this area.

    I personally OC my GeForce ti200 to ti500 speeds, it saved me $200 when i bought it. The same goes today, you can buy a Radeon 9500 pro and with a BIOS change you can make it a 9700 PRO, for a major difference in price. Now you can't tell me that's not worth it, for a free download and a little time you can save yourself $150. And trust me, you will see the difference in the benchmarks and in real-world usage.

    If you say otherwise, you just flat out don't know what you are talking about, so you should stop flaming others.
  13. Rick

    Rick TechSpot Staff Posts: 6,304   +52 Staff Member

    Phantasm has an extremely good point that I have neglected, as I've tried to argue in favor of overclocking...Stability.

    Even though it goes against my positive stance for overclocking, it CAN and WILL cause stabilty problems if not properly done. But nothing permanent... The risk is still there though and if outrageously done - May cause data loss or other unsightly things to happen.

    So if you want to argue overclocking, ISS, then stability might be a solid argument on your behalf.
     
  14. Phantasm66

    Phantasm66 Newcomer, in training Posts: 6,504

    I am sure I remember reading an article that the P75, P90 and P100 were basically EXACTLY the same piece of silicon, except that they had different writing on them!
  15. Justin

    Justin Newcomer, in training Posts: 1,595

    Sort of, phant, but not entirely.

    Basically, Intel took chips, and ran them all at 120mhz. Whichever parts were stable, they labeled as 120 and sold them off. The remaining they brought down to 100mhz and clocked them at that speed. Whichever were stable they sold as 100, then brought the rest down to 90. Et cetera, so on and so forth.

    The reason why so many 75- and 90- mhz parts reached 120 with just stock cooling and possibly a HSF (typically you can run a 200mhz pentium without a fan, or a 233mhz pmmx without a fan) is that they also had to sell a certain number of parts too. If they had too many 120mhz parts they could just label them as 75 and sell them off.

    However, chip manufacturers were doing this loooooooooong before Intel did it with the 2nd generation pentiums (75-200). Intel, AMD, Cyrix, Evergreen, IBM, etc, were all doing it with 386 and 486 parts as well. Overclocking saw some small light with 386 sx/dx parts and a little bit more with the 486, however it REALLY started to show its potential when the 486 DX2 hit the market. With this fangled thing called "multipliers", you could manipulate a CPU speed without having to also manipulate your bus speed, thus increasing your chance of a successful overclock plus actually offering more potential speeds. Most 486DX boards operated at 25, 33, 40, or 50mhz. 50mhz, however, was fairily uncommon and often unstable. The MOST common speeds were 25 and 33. So now that you have a part designed to run, say, at 25mhzx2, why not put it at 33mhzx2 ? Tada, you just came close to doubling your systems power at no cost to you.

    Don't think it was just intel doing this - AMD did/does the exact same thing and I am sure CPU manufacturers are still doing it today. In fact, overclocking would almost be non-existant if the chips they sold reached their maximum at the rated speeds anyways.
  16. Mictlantecuhtli

    Mictlantecuhtli TechSpot Evangelist Posts: 4,916   +9

    Many have said here that overclocking requires good and expensive ($50 something) heatsinks and things like that.

    That may be true for AMD users, but P4s can be overclocked a lot with their default heatsinks. I ran P4A 1.6 @ 2.4 with its default heatsink - that's 50% overclocking!
  17. acidosmosis

    acidosmosis TechSpot Chancellor Posts: 1,574

    overclock

    Yea, but you gotta realize, us users with Athlons running at say 1.8ghz, are running easily at the speed of P4 2100 gig users. Ah, the glory of Athlon.
  18. Arris

    Arris TechSpot Evangelist Posts: 4,508   +81

    Now we have 1700+ Tbred B running at 2400+ speeds with reasonable heatsinks... Athlon Rulez etc...
    The 1.6a P4 is one of the best overclocking CPUs for sometime to be honest. :)

    Welcome acidosmosis :wave:
  19. Mictlantecuhtli

    Mictlantecuhtli TechSpot Evangelist Posts: 4,916   +9

    Let's not start this AMD vs Intel again. I was just saying that expensive cooling isn't always required.
  20. Arris

    Arris TechSpot Evangelist Posts: 4,508   +81

    But you have to admit AMD are best!!! ;) (j/k)
    I don't want to get into the AMD vs Intel thing either.
  21. ---agissi---

    ---agissi--- TechSpot Paladin Topic Starter Posts: 2,382   +15

    The thing about Pentium4s is a 2.4Ghz P4 has the same performance as a 2ghz AthlonXP. As for o/c'ing, before I thought I'd go ahead and put forth. As for now, I'll stick with what I bought.








    AMD
  22. iss

    iss TechSpot Chancellor Posts: 2,896

    I dont see the above as an example of overclocking. the reason being that the 9500 pro and the 9700 pro are the EXACT same chip. the bios flash merely removes the artifically placed limitations that prevent the 9500 Pro from running at 9700 pro speeds. you would be "overclocking" only after you surpassed the 9700 clock speeds.

    the same is not UNIVERSALY true of processor chips. a 1600 palamino is not the EXACT same chip as a 2400 Tbred.
  23. ---agissi---

    ---agissi--- TechSpot Paladin Topic Starter Posts: 2,382   +15

    true ISS, but a 1600 pal. is the same chip as a 1700 pal. its just when the 1700 came out it worked stably as a 1700, the 1600 didnt, thats cuz its a 1600, not a 1700....all the chips are made exactly the same way, with the same stuff....its just some work better than others.;)
  24. Phantasm66

    Phantasm66 Newcomer, in training Posts: 6,504

    Yeah, lets NOT start this again.

    I think that AMD presents a total bargain, a CPU at prices that you can afford.

    Maybe a P4 has that little bit extra - but hey its too damn expensive!

    Both have their merits. Like Windows and Linux. Man, in the days past it was alpha vs i386. Lets just leave it at that.

    Computers are cool. End of story. If you are happy with your computer, with what you are doing with it, and what it is doing for you, then leave it alone.
  25. ---agissi---

    ---agissi--- TechSpot Paladin Topic Starter Posts: 2,382   +15

    thats the thing phant, people dont wanna leave it alone, they want more, because they can get more....thats just what overclocking is :D
Topic Status:
Not open for further replies.


Add New Comment

TechSpot Members
Login or sign up for free,
it takes about 30 seconds.
You may also...


Get complete access to the TechSpot community. Join thousands of technology enthusiasts that contribute and share knowledge in our forum. Get a private inbox, upload your own photo gallery and more.