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

which is better of these two brand new bartons?

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

cujo

Member
Joined
May 29, 2003
Location
Ontario, Canada
i currently have had the chip in my sig. just ordered another one for a friends rig that i'm building. pretty sweet computer actually. just wondering what's better? the one in my sig? or this new one which is a AQXFA 0334RPMW. that's right AQX"F"A. haven't heard a thing about these yet.

oh yea one more thing. the one in my sig is a brown pcb but this one is a green pcb.

now rather than taking my chip out and testing this in my machine i'll be testing it in the asus a7n8x deluxe rev2 that i'll be building it in. i'm guessin it won't make a difference which board i am using if i just use multipliers. i am building this other system with the new silent boost so i'll get a good idea of it's value as a hsf as well.
 
manufactured in the same week - I would say the only difference you will see will be down to the manufacturing process variation - hence pretty limited.
 
Akira283-IGN said:
If there is a difference, the newer one is better.

Not allways. I can be so that a newer process could go higher with low voltages, but hate high voltages, so the maximum oc can be lower. Good for somebody who is aiming for silence, and not using watercooling or noisy aircooling.
 
dropadrop said:


I can be so that a newer process could go higher with low voltages, but hate high voltages, so the maximum oc can be lower.

This is complete nonsense and you have no proof, factual or logical, to back this up. Newer is better.
 
well do you have any proof to back the whole newer is better thing?

some older weeks do much better than the newer ones

i remeember back in the day with the 1600 pallys a few older weeks would get hundreds of mhz higher than the newer ones

same thing with the 2100+ cpus

and same with the barton 2500 as well.... the 0330's seem to be the best week so far and that isnt the newest week....

whats the moral of the story? overclocking involves a great deal of luck.... some steppings / weeks do great and some do bad. there is ZERO way to tell without testing it out yourself
 
Copied/pasted for the thousandth time:

Weeks, sSpecs and steppings are practically useless. This is because not every CPU from every stepping and week is overclocked, so it's impossible to know the best overall. People only know which ones seem to overclock the best. Overclocking is unpredictable because a CPU from a popular stepping/sSpec/week might overclock poorly, and likewise a CPU from an unpopular stepping/sSpec/week might do very well. Less than 5% of CPUs are actually overclocked, so no one can argue that one type overclocks more than another.

Another factor is user error. A certain CPU might be capable of 3.3GHz on air-cooling, but if the user isn't a very knowledgeable overclocker, or one or more of his components are sub-standard, he may only reach 3GHz. This user then decides to go on some forum and write about his "crappy stepping that only reaches 3GHz," when the fact is that the CPU is capable of more. This is why no stepping will guarantee a good overclock, or indicate the speed you will get. Every CPU is different, so the only way to find out how far you can overclock is to try it and see.

As time goes on, manufacturing techniques improve, which is why AMD and Intel are able to increase clock speeds. The engineering improvements from one stepping to a newer one is reason for the increase in overclocking potential. Some overclockers believe that steppings have some hidden meaning for overclocking, but that is just a myth.

Steppings have no affect on overclocking - they simply signify that some type of manufacturing refinement was made. Newer equals better in the overclocking world, but some ignorant overclockers like to argue that older steppings do better than newer ones. They base these theories off of only a few results, and fail to understand that truth that intelligent overclockers already know.

Some people even argue that some older production weeks are better than newer ones for overclocking. False! Newer CPUs based on the same stepping always have more potential than older ones due to the constant rate of improving technology. Weeks themselves do not indicate overclocking potential. I see so many false arguments like "week 31 is better then week 35, but worse than week 24." All you need to do to get the most overclocking potential is to simply buy a new CPU. If you can find a place to get a guaranteed week, great, that way you can simply choose the newest CPU possible. Just don't be fooled by myths that older weeks are better.

As a general rule, the newer the CPU, the more overclocking potential it has.
 
I'd like to see what the new stepping will do. Arguments of 'newer is better' aside, make sure you let us all know what it tops out at ;)
 
indeed too much of this newer date is better.. its all BS!

alot of TBRED A DLT3C has a NEWER DATECODE than many TBRED B DLT3C that ppl have here and will get OWNED by any TBRED B :eek:
 
Akira283-IGN said:


This is complete nonsense and you have no proof, factual or logical, to back this up. Newer is better.

Just as an example, when the dlt3c processors came out, everyone was jumping on them. They do well with low voltages, but they often stop scaling after a point. Soon after they where released, people where getting higher overclocks with dut3c processors by pumping 2-2.1v through them, while the dlt3c's would'nt really benefit from over 1.9v even when using a prometia for cooling. At the time people where not getting overclocks of over 2.8ghz out of the dlt3c's, while the dut3c's where going closer to 3ghz.

I don't think there is any rule that applies 100%, and naturally often the newer cpu's do overclock better. When the manufacturer refines the manufacturing process to get better yields, it often results in a higher overall overclock for us.

In the example above, they had managed to refine the process in a way that they get better yields, as the processors scale higher with default voltage. For us it was'nt as good as we where ready to pump alot more voltage through them then what they where designed for.

I would'nt count the newer t-bred a's into the equation, as I don't think amd has manufactured t-bred a based cores for some time now. It must be old processors that have just been packaged later?
 
ehehe why are you wasting your breath explaning something everyone on the whole forum understands knows and accepts.

He's beening trying to tell the whole forum that something that is more or less been proven with stats from diff stepping cpus is false.. Hard headed if you asked me
 
Back