PDA

View Full Version : Help Me Decide The Final Moments Before I Purchase!


OC-Master
10-15-03, 09:46 PM
In under 15 hours, I'll be ordering my new AMD AthlonXP CPU and I need to decide whether its going to be the 1800+ or 2100+ Model.

The question is simple, 1800+ or 2100+.. MONEY IS NO ISSUE since the difference is $15 LOL

1800+ is JIXIB

2100+ is JIXIB

I'm aiming towards the 2100+ because of the 13X~24X multipliers being available without the need of any modding. My motherboard only goes to 12.5X and my FSB doesnt like anything past 144MHz.

But.. Does the 1800+ reach higher MHz or vice versa?

Overclocking potential? Anyone have input?

Thanks everyone in advance for any input.

P.S - I'm getting a P90F ViewSonic monitor too :D Tomorrow will be a good day hopefully.


OC-Master

Gautam
10-15-03, 09:51 PM
No way! Is it really time to say goodbye to the ol' T-Bird? My hypothesis is that the branded speed is irrelevent, i.e. a 2100+ JIXHB DLT3C 0330 XPMW would clock exactly as well as an 1800+ of the same stepping. Since the 2100+ will be more convenient for you, I'd say its the better choice.

OC-Master
10-15-03, 09:53 PM
Originally posted by Gautam
No way! Is it really time to say goodbye to the ol' T-Bird? My hypothesis is that the branded speed is irrelevent, i.e. a 2100+ JIXHB DLT3C 0330 XPMW would clock exactly as well as an 1800+ of the same stepping. Since the 2100+ will be more convenient for you, I'd say its the better choice.


Nice..

very quick responce.. thanks for the reply..

I'll take that into thought for my purchase :)

And yes, its time that my AXIA 1000C Athlon retire..

All this MP3 compression is wasting my time and I need to cut rendering times in half somehow.

I'm hopping to hit 2.50GHz but I dunno how my chances are.


OC-Master

Charloz24
10-15-03, 10:00 PM
Your chancea are pretty good :)

cack01
10-15-03, 10:03 PM
I would take the 1800+.
1. its cheaper. $15 still gets you a movie ticket.
2. 2.5ghz= 12.5*200

You are most likely going to want to run a FSB higher than 200, so you are not going to need a multiplier higher than 12.5. Although you stated that 144FSB seems to be the max for your mobo so therefore the 2100+ is probably the safer route. Maybe you should get a new mobo first, as I suspect you will end up not using this chip to its fullest.

c627627
10-15-03, 10:12 PM
If they're both JIXIB and you need 13 and above multipliers, only Abit and Soltek nForce2s (as far as was reported here) can post higher multipliers on below 2100+ CPUs. I think all VIA mobos can too, but they don't have a PCI lock so there goes that advantage.

You do have an option of a wire trick on the 1800+ but it can be a hassle...

hitechjb1
10-15-03, 10:13 PM
Are you sure the 2100+ JIXIB is a DLT3C, not a DUT3C?

I would prefer a DLT3C over a DUT3C, regardless the PR number. Lower voltage rating mostly means lower transistor threshold variation, hence higher overclocking potential at same voltage.

Deathknight
10-15-03, 10:22 PM
Yeah got for an 1800+ DLT3C. The price is right and it ocs great. If you need the higher mults the pin mod is really not that big a deal and its easily reversable.

OC-Master
10-15-03, 10:22 PM
Ok, I didnt even know a 2100+ DLT3C even existed. I thought it was only on 1800+ and 1700+ CPUs??

I believe both CPUs are of Week 36 if that helps at all.


OC-Master

Gautam
10-15-03, 10:22 PM
Originally posted by hitechjb1
Are you sure the 2100+ JIXIB is a DLT3C, not a DUT3C?

I would prefer a DLT3C over a DUT3C, regardless the PR number. Lower voltage rating mostly means lower transistor threshold variation, hence higher overclocking potential at same voltage.

You can't compare the stock voltage of a 1.53ghz processor with that of a 1.73 ghz one directly. Even if the 2100+ is rated DUT3C, it is only because of the increased speed.

c627627
10-15-03, 10:28 PM
Proof that there are no 2100+ DLT3Cs:
official AMD voltage chart:
Copy-paste without "
"http://139.95.253.214/SRVS/CGI-BIN/WEBCGI.EXE/,/?St=24,E=0000000000110206233,K=7607,Sxi=4,Case=obj (1224)"

hitechjb1
10-15-03, 10:29 PM
I consider a 1.5 V rated DLT3C would have better overclocking potential than a 1.6 V rated DUT3C, based on the intrinsic transistor characteristics (potentially lower channel length and lower threshold variation). These difference has been discussed quite a few times before, I think.

So do you know whether the 2100+ is a DUT3C or a DLT3C? From the model 8 datasheet, there is no 2100+ DLT3C.

Gautam
10-15-03, 10:40 PM
Yes, you have discussed this topic very thoroughly, although I thought you were showing the difference between a 1700+ DUT3C and DLT3C, not 2100+ vs. 1700+. Its not fair to say, for example that a 2700+ will overclock worse than a 1700+ due to its stock voltage being higher. Its stock voltage is higher because its stock speed is higher. Similarly, its not fair to compare the stock voltage of a 2100+ with that of a 1700+. The voltage of the former needs to be increased concurrently with the stock speed increase. Even the new 2100+'s could possibly have very short transistor channel lengths.

hitechjb1
10-15-03, 10:40 PM
In case for those who are interested, this is what I think:

Originally posted by hitechjb1

Lower voltage, shorter transistor channel length, lower transistor threshold voltage and Tbred B 1700+/1800+DLT3C

The Tbred B 1700+ can perform so well in general, as reported by so many people, such coincidence is not just by luck. I think it has something to do with its intrinsic transistor properties, resembling some future trends for silicon scaling into future generations, namely, lower voltage, shorter transistor channel length, faster transistors for the good, but higher leakage current for the bad.

There are process variations of a given silicon manufacturing process, as in any manufacturing process. As a result, the intrsinic silicon proporties, such as transistor channel length and width, gate oxide thickness, silicon carrier doping, transistor threshold voltage, leakage current, random manufacturing defects, ..., of a chip in a silicon wafer can vary to certain extent (sigma variation). As further scaling down, statistical variation comes into play, i.e. nearby transistors with the same design attribute in the same chip/wafer can even behave differently.

A more interesting question is what the implications of these wafer properties of lower threshold voltage and shorter channel length due to process variation of a manufacturing process are, as I suspect for the the Tbred B DLT3C. It is being rated at lower Vcore but it can run faster than other Tbred B at same voltage. Even it is manufactured with 0.13 micron like other Tbred B, it is effectively behaving like a chip with less than 0.13 micron, resembling the future generation trend.

As the transistor size (channel length) of future generations of silicon chips are scaled down to, e.g., 90, 65, 45, ... nano-meter (nm) (e.g. Hammers are 90/130 nm SOI, TBred B is 130 nm, Palomino is 180 nm), the supply voltage, transistor channel length and threshold voltage will be lowered accordingly. Even the supply voltage is lower, the transistors run faster, both current and power density also increase (actual trend). As the transistors are scaled down, logic gate delay decreases, both the active power density (W/cm^2) and the passive leakage power density (from both gate and subthreshold leakage) increase.

The passive leakage current component increases at an even faster pace than the active current, posing problems on cooling and power dissipation for future generations of chips. If this trend continues, the high passive, standby leakage current will lead to high power drawn and high idle CPU temperature, compared to today's CPU, even when the system is idle and the CPU is not under heavy load.

For more details about the low voltage Tbred B 1700+, refer to
Why the 1700+ can run so fast at low Vcore? (http://forum.oc-forums.com/vb/showthread.php?s=&postid=1734079#post1734079)

brennan77
10-15-03, 11:03 PM
I agree. Go with the DLT3C rather than the DUT3C. I had a JIUHB/DUT3C 2100 that would take over 1.925v to hit 2200 with Prime95 stability. My JIUHB/DLT3C 1800 hits 2400 at 1.85v and 2450 on a cool night.

I don't think you can directly conclude an absolute rule here, but I'd go with the DLT3C.

OC-Master
10-15-03, 11:51 PM
Does anyone have experience overclocking the JIXIB 2100+??


OC-Master

Gautam
10-15-03, 11:55 PM
Okay, perhaps I wasn't clear enough, but there's an obvious reason why the 2100+ is rated at a higher stock voltage. It has a higher clock speed. There's nothing special about its lower stock speed. That's like saying an overclock at 2.0ghz using 1.6v for example is better than an overclock at 2.5ghz using 1.95v because the amount of voltage being used is less in the former. The two cannot be compared directly. OC-Master, the very very best thing you could do for yourself and the rest of us is buy both of them and test them out to see how they do. ;) Maybe then we could get an accurate picture of how steppings do against one another, and you could know for sure which one is the better proc! :D Both of them are going to be killer overclockers for sure, so I wouldn't think you'd have a problem in selling one...

OC-Master
10-16-03, 12:07 AM
Originally posted by Gautam
Okay, perhaps I wasn't clear enough, but there's an obvious reason why the 2100+ is rated at a higher stock voltage. It has a higher clock speed. There's nothing special about its lower stock speed. That's like saying an overclock at 2.0ghz using 1.6v for example is better than an overclock at 2.5ghz using 1.95v because the amount of voltage being used is less in the former. The two cannot be compared directly. OC-Master, the very very best thing you could do for yourself and the rest of us is buy both of them and test them out to see how they do. ;) Maybe then we could get an accurate picture of how steppings do against one another, and you could know for sure which one is the better proc! :D Both of them are going to be killer overclockers for sure, so I wouldn't think you'd have a problem in selling one...

Ironically, I'm getting a 2100+ for my self and one for my sister. Our system setups are nearly identical.
After reading a few posts, I now think its the wise choice to go with the 2100+ because of the remapped multipliers and the fact i dont have to mod no more LOL My motherboard is volt modded and bios hacked, should do the trick.

Maybe now my Enermax 460Watt PSU will get a run for its money?


OC-Master

hitechjb1
10-16-03, 12:14 AM
If I have to buy for myself, I would get the 1800+ DLT3C, unless the 2100+ is a DLT3C (not likely), for the reason I mentioned in earlier posts. The multiplier shifting is a matter of putting jumper wire in the socket carefully, and is reversible. Further it is $15 cheaper.

tediber
10-16-03, 12:30 AM
if money is really no question ...... you should be getting a barton 2500+ instead as newer steppings of bartons are out 033x weeks

c627627
10-16-03, 12:33 AM
Hang on there OC-Master, Abit mobo?

Maybe it reads all multis on all processors like Abit nForce2. If this is true, you wouldn't need any wire tricks so you could get either one. Get both and see which one's better... do it for your pals here at the forum... :)

Gnufsh
10-16-03, 07:58 AM
I doubt his board supports the higher multis, it is a KT133A board after all.

brennan77
10-16-03, 08:30 AM
Originally posted by Gautam
Okay, perhaps I wasn't clear enough, but there's an obvious reason why the 2100+ is rated at a higher stock voltage. It has a higher clock speed. There's nothing special about its lower stock speed. That's like saying an overclock at 2.0ghz using 1.6v for example is better than an overclock at 2.5ghz using 1.95v because the amount of voltage being used is less in the former. The two cannot be compared directly. OC-Master, the very very best thing you could do for yourself and the rest of us is buy both of them and test them out to see how they do. ;) Maybe then we could get an accurate picture of how steppings do against one another, and you could know for sure which one is the better proc! :D Both of them are going to be killer overclockers for sure, so I wouldn't think you'd have a problem in selling one...

Correct me if I'm wrong, but aren't there actual differences between the DLT3C and DUT3C apart from the simple Vcore setting? Isn't that why the DLT3C can hit higher speeds at less voltage? I was under the impression that there was more to it than what you propose.

Again, I thought I was gonna have a great overclocker in a 2100 JIUHB DUT3C. The JIUHB DLT3C's were kickin butt afterall. But it didn't turn out like I expected and several others with the same CPU had the same results. I'm just saying that it's a possibility that the JIXIB DUT3C will not overclock like it's DLT3C brother.

Col_Loki
10-16-03, 09:52 AM
On the Abit unlocking the higher level multi's subject..... i know it definetly unlocks them on the Kt400 chipsets ie the KD7 boards.
Mabye they always had that on their boards??

[edit] I think the NewEgg bartons are Aqzfa steppings which are set to be brilliant overclockers..... extra cache aswell.............. worth thinkin about?

OC-Master
10-16-03, 08:24 PM
Originally posted by tediber
if money is really no question ...... you should be getting a barton 2500+ instead as newer steppings of bartons are out 033x weeks


But then I would have to take the chance and attempt the multiplier mod. This also has to be for my sisters board and she wont let me mod it at all.

The only thing she'll let me do is plug the CPU in and overclock to a maximum of 2.40GHz regardless.


OC-Master

c627627
10-16-03, 08:32 PM
phiew 2.4, that all?

:) That's actually above average and not that easy to do.
2.4 is a PR rating in the 3000+s.

OC-Master
10-16-03, 09:18 PM
Originally posted by c627627
phiew 2.4, that all?

:) That's actually above average and not that easy to do.
2.4 is a PR rating in the 3000+s.

Well, her board aint voltage modded so the thing actually maxes out at 1.85vcore.

My board is voltage modded so I can go as high as I want. :D

My goal is 2500MHz


OC-Master

c627627
10-16-03, 10:58 PM
OC-Master, may I kindly ask you where you got the idea that 2500 AMD MHz is within easy reach?

A particularly good nForce2 mobo with top notch name brand power supply and top of the line cooling (both case and CPU) is required 9/10 for 2500 MHz and that's assuming the most important factor is taken care of (luck).

23xx Mhz overclock is the average. Anything above that is a cause for celebration and it almost always depends on additional components not just the CPU.

Having said that, we're rooting for you all the way and are anxious to find out just how much it really all depends on the CPU itself. Good luck buddy.