-
Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!
Thor(n)ton - please post if you know where to get one
- Thread starter c627627
- Start date
- Joined
- Jan 18, 2003
- Location
- Murphy, Dallas.
I got 2200+ today at FRY's for $75. It run cooler than Tbred. I havent try to enable anther 256K L2 cache. It not a good O/C chip the max I can go with air cooling is 2.4G Vcore 1.9V but unstable, just got into Windown and restart, temp i got at the time system restart 42C. Stable @ 11.5 * 200 with Vcore 1.875. I think with better cooling it can go better!
Stepping: AQZFA 0339 UPMW
Board: Epox 8RGA+
Mem : Samsung PC2700 (it can go 430MHZ)
VCard: MSI TI4200-64M
HSF : Alpha PAL8045U + TT Smart II
At 11.5 * 200 temp never go over 34C. Cooler than 2100+ i have.
I think the best way to go still 1700+ or 1800+ ( if i have a change).
Stepping: AQZFA 0339 UPMW
Board: Epox 8RGA+
Mem : Samsung PC2700 (it can go 430MHZ)
VCard: MSI TI4200-64M
HSF : Alpha PAL8045U + TT Smart II
At 11.5 * 200 temp never go over 34C. Cooler than 2100+ i have.
I think the best way to go still 1700+ or 1800+ ( if i have a change).
OP
- Joined
- Feb 18, 2002
- Thread Starter
- #43
As soon as I find out the CPU limits brennan77, I'll mod this Thorton thanks to you.
Guess what, no multiplier between 8 and 13 will post with any FSB exept for the 12.5 (default 2000+ multiplier) on Epox 8RDA+ nForce2 (!)
Guess what, no multiplier between 8 and 13 will post with any FSB exept for the 12.5 (default 2000+ multiplier) on Epox 8RDA+ nForce2 (!)
- Joined
- Aug 8, 2001
- Location
- New Orleans
So you have no multiplier adjustment whatsoever? Odd. And you're welcome. It's for the cause of seeing what these things will do. Plus, you've helped me out on occasion with your productive posts. So it's really a thanks to you. 
Awww... Now we're getting all warm and fuzzy in here.
But anyway....
How far can you usually push your FSB with your RAM and what not? That might be the limiting factor here without multiplier adjustments.
Awww... Now we're getting all warm and fuzzy in here.But anyway....
How far can you usually push your FSB with your RAM and what not? That might be the limiting factor here without multiplier adjustments.
c627627 said:
I also got the same 92 mm fan, w/ a SLK-947U.
You have to turn the speed dial up (fully clockwise), otherwise at the lowest dial setting, the fan hesitates to start.
First impression, it is quieter than the TT SFII at full blown, but also it does not put out as much CFM as the TT SFII at max speed (as expected from spec). So it is not as versatile as a TT SFII for max overclocking.
Will also test the 947U w/ a 92 mm Tornado.
OP
- Joined
- Feb 18, 2002
- Thread Starter
- #46
At newegg it's $5.99 instead of SVC's $11.99. Just so you all
know.
and here's the result of my test with this fan:
All default settings
1.) Enermax 92x25mm UC-9FAB fan NOT connected: All normal.
2.) Connected to FAN3 slot, system freezes at black BIOS screen.
3.) Connected to either FAN2 or FAN1, system does not switch on at all (!). All normal after the fan is disconnected.
4.) Thanks to hitechjb1's suggestion: "You have to turn the speed dial up (fully clockwise), otherwise at the lowest dial setting, the fan hesitates to start."
The system boots without a problem after I did what hitechjb1 said.
How can this affect the entire system not switching on??
Good thing I looked at this thread just as I was beginning to start my investigation into why the fan was causing this. Thanks for the hint hitechjb1.
know.
and here's the result of my test with this fan:
All default settings
1.) Enermax 92x25mm UC-9FAB fan NOT connected: All normal.
2.) Connected to FAN3 slot, system freezes at black BIOS screen.
3.) Connected to either FAN2 or FAN1, system does not switch on at all (!). All normal after the fan is disconnected.
4.) Thanks to hitechjb1's suggestion: "You have to turn the speed dial up (fully clockwise), otherwise at the lowest dial setting, the fan hesitates to start."
The system boots without a problem after I did what hitechjb1 said.
How can this affect the entire system not switching on??
Good thing I looked at this thread just as I was beginning to start my investigation into why the fan was causing this. Thanks for the hint hitechjb1.
c627627,
the enermax fans give weird rotation results to the board, and the board doesn't start. The alternative is to unplug the yellow wire. I saw this with my 8RDA+ and an NF7-S.
By the ay, I have a thorton, but not to impressed with it. It does 2.4 Ghz at 1.85v with an SLK800 + 4000 rpm 80mm fan, but eventually it may give an error. 1.875v seems to be stable. I had a 2500+ that do 2.4 with just 1.8v, and the difference in price between them (here in Portugal, about 15 €) doesn't justify the thorton.
The mod worked ok.
the enermax fans give weird rotation results to the board, and the board doesn't start. The alternative is to unplug the yellow wire. I saw this with my 8RDA+ and an NF7-S.
By the ay, I have a thorton, but not to impressed with it. It does 2.4 Ghz at 1.85v with an SLK800 + 4000 rpm 80mm fan, but eventually it may give an error. 1.875v seems to be stable. I had a 2500+ that do 2.4 with just 1.8v, and the difference in price between them (here in Portugal, about 15 €) doesn't justify the thorton.
The mod worked ok.
Well, this looks promising so far - such a new chip surprises me a lot as I thought that these would have been discontinued by now and replaced by T-Breds/Bartons. I can't wait to see your results with, and without cache on this baby!
Should do 2.4 without too much difficulty!
Should do 2.4 without too much difficulty!
c627627 said:At newegg it's $5.99 instead of SVC's $11.99. Just so you all
know.
and here's the result of my test with this fan:
All default settings
1.) Enermax 92x25mm UC-9FAB fan NOT connected: All normal.
2.) Connected to FAN3 slot, system freezes at black BIOS screen.
3.) Connected to either FAN2 or FAN1, system does not switch on at all (!). All normal after the fan is disconnected.
4.) Thanks to hitechjb1's suggestion: "You have to turn the speed dial up (fully clockwise), otherwise at the lowest dial setting, the fan hesitates to start."
The system boots without a problem after I did what hitechjb1 said.
How can this affect the entire system not switching on??
Good thing I looked at this thread just as I was beginning to start my investigation into why the fan was causing this. Thanks for the hint hitechjb1.
c627627 said:
1. See next post.
2. For the TT SFII, you have to take out that SMALL JUMPER PLUG that shorts two pins, in order for the manual adjustment to be able to function. When that two pins are shorted, the fan is operating in the sensor mode and the fan knob is disabled.
3. For the Enermax 92 mm fan, when the knob is fully counterclockwise, it is either OFF, or at a setting at very slow speed and it may not spin due to friction (don't know for sure which is the case). One has to turn up the knob a little bit, in order for it to spin, ...
For some motherboards, if the auto fan speed protection is on, the motherboard would not start if the fan is OFF or the speed is below certain RPM. Hence in your case, the system did not boot.
This is not very user friendly in my opinion, since the fan come in such default setting and many may mistake that something is defective. Imagine one is building a new system with everything new and this fan, and may have a hard time to figure out whether it was the motherboard, fan, bios, ... scary !!
Last edited:
Regarding to the 2.3 or 2.4 GHz CPU w/ 256 or 512 KB L2:
1. For both Tbred and Barton, a CPU at 2.4 GHz would have more raw power than a 2.3 GHz CPU, about 4.3% which would show up in CPU benchmarks such as Dhrystone, Whetstone for integer or floating computation respectively, .... These CPU benchmarkings reuqire small amount of memory that can be fit in the cache (both 256 KB or 512 KB) to test the CPU, so cache size would not make a difference. That is why a Tbred w/ 256 KB cache and a Barton w/ 512 KB cache both have the same IPC, which measures the amount of CPU instructions executed per cycle for a given set of chosen prgrams.
In general, for many programs whose inner loop and data cannot be fit into a smaller L2 cache (shared w/ other programs and threads) over time (many cycles), the usefulness of larger cache will come in to improve the overall performance of CPU + cache + memory, since the probablility that the CPU needs to get data from memory (cache miss) is less for larger cache, and getting data from memory requires many CPU cycles. For a CPU w/ a bigger cache running real life programs, the chance to find the next few instructions and data for a program in the cache is higher. Such improvement levels off when cache size is increaed beyond certain size (>> 512 KB).
From benchmark results of many programs, the large 512 KB L2 gives peformance improvement ranging from 0 - 10+%. So in simple terms, take 5% as average without going into which specific programs are being compared. That is why for CPU at 2.4 GHz, 100-120 MHz (4-5%) is chosen as the tradeoff measure between the slightly faster 256 KB Tbred B, and the larger 512 KB L2 Barton. To evaluate more precisely, one would need to benchmark a particular program or set of programs specifically.
1. For both Tbred and Barton, a CPU at 2.4 GHz would have more raw power than a 2.3 GHz CPU, about 4.3% which would show up in CPU benchmarks such as Dhrystone, Whetstone for integer or floating computation respectively, .... These CPU benchmarkings reuqire small amount of memory that can be fit in the cache (both 256 KB or 512 KB) to test the CPU, so cache size would not make a difference. That is why a Tbred w/ 256 KB cache and a Barton w/ 512 KB cache both have the same IPC, which measures the amount of CPU instructions executed per cycle for a given set of chosen prgrams.
In general, for many programs whose inner loop and data cannot be fit into a smaller L2 cache (shared w/ other programs and threads) over time (many cycles), the usefulness of larger cache will come in to improve the overall performance of CPU + cache + memory, since the probablility that the CPU needs to get data from memory (cache miss) is less for larger cache, and getting data from memory requires many CPU cycles. For a CPU w/ a bigger cache running real life programs, the chance to find the next few instructions and data for a program in the cache is higher. Such improvement levels off when cache size is increaed beyond certain size (>> 512 KB).
From benchmark results of many programs, the large 512 KB L2 gives peformance improvement ranging from 0 - 10+%. So in simple terms, take 5% as average without going into which specific programs are being compared. That is why for CPU at 2.4 GHz, 100-120 MHz (4-5%) is chosen as the tradeoff measure between the slightly faster 256 KB Tbred B, and the larger 512 KB L2 Barton. To evaluate more precisely, one would need to benchmark a particular program or set of programs specifically.
Last edited:
OP
- Joined
- Feb 18, 2002
- Thread Starter
- #51
No go on the 512 mod.
It could be because I have a green Thorton and all the mods talk about brown Thortons. Green ones may require additional steps like this:
http://www.ocprices.com/?rev_id=184
All I did was fill the L2 laser pit with nonconductive paint. (Even tried filling it with white out). I was careful for the non conductive paint not to get on the bridge dots.
Then I connected the dots on each side of the pit with conductive compound which went straight over the recently filled laser pit.
Do you all think the only way to mod a green thorton is to do scraping like this:
http://www.ocprices.com/?rev_id=184
It could be because I have a green Thorton and all the mods talk about brown Thortons. Green ones may require additional steps like this:
http://www.ocprices.com/?rev_id=184
All I did was fill the L2 laser pit with nonconductive paint. (Even tried filling it with white out). I was careful for the non conductive paint not to get on the bridge dots.
Then I connected the dots on each side of the pit with conductive compound which went straight over the recently filled laser pit.
Do you all think the only way to mod a green thorton is to do scraping like this:
http://www.ocprices.com/?rev_id=184
- Joined
- Aug 8, 2001
- Location
- New Orleans
Yeah, you might have to do some scraping.