>>>Abit IC7 875P Vs Albatron 845PE<<<

oldfart

Lifer
Dec 2, 1999
10,207
0
0
Finally got the IC7 up and running and thought I'd put up some info. Its a nice board with everything I need and a little more. 8 USB ports is nice as is built in TI Firewire. One less PCI card. I didn't try out the onboard sound since I have a TBSC sound card. The board is big. Much larger then the Albatron 845PE that came out.

I'm happy to report that the NB fan is actually very quiet despite the ~6500 RPM fan! It is nowhere near as loud as the fan on my Gigabyte Radeon 9700 Pro. That will be fixed soon with a Zalman ZM80A-HP passive cooler.

The layout of the board is very nice. Everything fits well. Cables are easy to dress and keep out of the way of airflow. I've always liked IDE ports that are "on edge" instead of facing up.

I started with the version 10 BIOS which overrides manual cas settings. The 11 corrects it but causes some SATA issues. I'm not using SATA, so it didn't affect me. The 11 BIOS provided a nice performance increase.

I've also put up benches of my 845PE running same FSB, CPU speed, but SC DDR 425 Vs. DC DDR 340 to show the performance increase.

As you can see, **IT IS NOT A LOT**!. You only get ~ 3 - 6 % of a boost going from a SC DDR setup using a ratio to a DC DDR rig. Much of the gain is from the ability to run lower cas timings. An 845PE is still a VERY GOOD rig for 533 MHz FSB P4s!

Here are some benches in the Thugsrook format (thanks)



Abit IC7 (i875P) vs Albatron PX845PEV Pro (i845PE)

--------------------------------------------------------------------------------

TEST SYSTEM
P4 2.4B @ 3.06 GHz 170fsb
Corsair XMS3200C2 2x 512 mb
Alpha 8942 - Panaflo L1A
Radeon 9700P - Cat 3.2 - stock speed - no AA/AF
Win2K SP3

TEST SPECS
ABIT IC7 i875P = 1.575vcore | AGP/PCI 68/34 | DC DDR 340 | cas 2-3-3-6 / 2-2-2-6
Albatron i845PE = 1.575vcore | AGP/PCI 68/34 | SC DDR 425 | cas 2.5-3-3-7


--------------------------------------------------------------------------------

BENCHMARKS

NOTE
IC7-11 = IC7 with BIOS ver 11 cas 2-2-2-6
IC7-10 = IC7 with BIOS ver 10 cas 2-3-3-6

> 3DMark2001 < default
IC7-11 = 15983
IC7-10 = 15711
Albatron = 15512


> UT2003 Demo < 1024x768x32, flyby - botmatch
IC7-11 = 219 - 78.9
IC7-10 = 215.2 - 77.3
Albatron = 212.6 - 75.6


> Quake III 1024x768x32
IC7-11 = 368
IC7-10 = 360
Albatron = 347

> Seroius Sam SE Crand Cathedral 1024x768x32
IC7-11
211.5 KB used for demo profile:
Originally recorded: 10777 frames in 96.9 seconds => 111.3 FPS average.
Without excessive peaks: 10439 frames in 91.8 seconds => 113.8 FPS average.
High peak: 249.0 FPS
Low peak: 71.6 FPS
Low sustained: 88.9 FPS (127 frames in 1.4 seconds)

IC7-10
206.7 KB used for demo profile:
Originally recorded: 10530 frames in 96.9 seconds => 108.7 FPS average.
Without excessive peaks: 10225 frames in 91.9 seconds => 111.3 FPS average.
High peak: 245.5 FPS
Low peak: 69.2 FPS
Low sustained: 84.2 FPS (66 frames in 0.8 seconds)

Albatron
203.5 KB used for demo profile:
Originally recorded: 10366 frames in 96.9 seconds => 107.0 FPS average.
Without excessive peaks: 9965 frames in 90.6 seconds => 110.0 FPS average.
High peak: 234.3 FPS
Low peak: 69.2 FPS
Low sustained: 84.1 FPS (65 frames in 0.8 seconds)


> TMPGEnc 2.53 MPEG-2 encode time (m:ss)
IC7-11 = 2:08
Albatron = 2:14

> LAME MP3 Decode MP3 -> Wave (m:ss)
IC7-11 = 2:27
Albatron = 2:28

 

maxSe

Golden Member
Aug 23, 2000
1,947
0
0
Cool~ Nice work oldfart. :):cool:

It's interesting that you didn't experience any performance/stability problem that THUGS had with 11 bios...
 

oldfart

Lifer
Dec 2, 1999
10,207
0
0
Not at all. As you can see, the 11 BIOS with the better cas timings increased performance as would be expected.
 

maxSe

Golden Member
Aug 23, 2000
1,947
0
0
I assume this is done with NBS 533, right? Were you getting same results/scores with NBS 800 with same cas timing? I think this is where the bug(?) with 11 bios occurs, and causes concerns with IC7 when used with P4 "C" chips... I don't mind using the 10 bios right now since I only have a "B" chip, and NBS533 is setting my timing to 2-2-2-6 anyway... :)
 

oldfart

Lifer
Dec 2, 1999
10,207
0
0
Originally posted by: maxSe
I assume this is done with NBS 533, right? Were you getting same results/scores with NBS 800 with same cas timing? I think this is where the bug(?) with 11 bios occurs, and causes concerns with IC7 when used with P4 "C" chips... I don't mind using the 10 bios right now since I only have a "B" chip, and NBS533 is setting my timing to 2-2-2-6 anyway... :)
Never bothered to try NBS 800. Didn't see any point of doing so. There is no reason to set NBS 800 on a 533 CPU. The bug is in the 10 BIOS, not 11. The 11 BIOS has a SATA problem, but fixes the cas problem. NBS 533 set mine to 2-3-3-6, so the 11 BIOS was a good option for me.

 

Mikki

Golden Member
Jun 13, 2002
1,488
0
0
Nice comparison oldfart, it seems the newer bios is much better...I like your tests....;):)
 

WarCon

Diamond Member
Feb 27, 2001
3,920
0
0
I just setup my friend's IC7 over this last weekend and was very impressed with the overall quality and features of this motherboard. Seems to be very stable. Managed to get his 2.4B SL6EF to 3.15ghz (175fsb) at 1.575vcore. (Will do 178fsb at 1.6vcore, but turned it down a bit for summer heat that will be coming).

What I thought was really nice, was his Gainward ti4200 (which I recommended) wouldn't overclock at all in his P4S333 even with a nice cooling kit on it. I felt pretty bad for him, but now in this IC7 it has wings. It hits 290/565 without a struggle, might be able to do more. This brought his 3dmark2001 scores up from 10k to over 14k like it should be. And the onboard sound is fantastic. It is much better than his old soundblaster 5.1 live gamer.
 

Duvie

Elite Member
Feb 5, 2001
16,215
0
71
Oldfart please touch on some of these things for us....


1) temps on board...cpu and sys...

2) board volting...undervolting/ or overvolting....

Looks nice though!!!!


I wonder was that the highest you could hit with that chip??? Isn't it true the DCDDR will give more benefit per each jump you increase in the fsb??? IE 3% now could be 4 to 5% maybe 5fsb later????


I agree my PE chipset is nothing to shake a stick at and I will keep it a bit longer...I think I may be more interested in some of the i865 mobos when they have a chance to mature a bit....
 

THUGSROOK

Elite Member
Feb 3, 2001
11,847
0
0
nice work oldfart :cool:

the problem is was having on the board was cause of the XMS LL ram i was using.....
IC7_10 cas2225 was a piece of cake and had great performance~
IC7_11 was also cas2225 but it also adds 4 turbo type settings~
all of those turbo settings refused to remain stable for benchmarks hence slower scores then the #10 bios~ :(

:)
 

motojeff

Member
Mar 21, 2002
115
0
76
Hey old - ahem - fart, you making me jealous!

I get 35 fps on cathedral in 16 bit color! u are 3x faster than my vc820, ti4200, pc800, 866...

Time to upgrade, but would like to hold on till Prescott! Droooool
 

oldfart

Lifer
Dec 2, 1999
10,207
0
0
Originally posted by: THUGSROOK
nice work oldfart :cool:

the problem is was having on the board was cause of the XMS LL ram i was using.....
IC7_10 cas2225 was a piece of cake and had great performance~
IC7_11 was also cas2225 but it also adds 4 turbo type settings~
all of those turbo settings refused to remain stable for benchmarks hence slower scores then the #10 bios~ :(

:)
I saw the new settings, but just left them all default. As you can see, 11 gave me a pretty large performance increase.
 

oldfart

Lifer
Dec 2, 1999
10,207
0
0
Originally posted by: Duvie
Oldfart please touch on some of these things for us....


1) temps on board...cpu and sys...

2) board volting...undervolting/ or overvolting....

Looks nice though!!!!


I wonder was that the highest you could hit with that chip??? Isn't it true the DCDDR will give more benefit per each jump you increase in the fsb??? IE 3% now could be 4 to 5% maybe 5fsb later????


I agree my PE chipset is nothing to shake a stick at and I will keep it a bit longer...I think I may be more interested in some of the i865 mobos when they have a chance to mature a bit....

I haven't really tried that hard to get more out of the CPU. 175 FSB on either board is not stable @ 1.6 Vcore. I wont go any higher. I haven't got down to trying 171 - 174 yet.

Lets see, Temp readings are off. Case temp seems pretty close, but CPU temp reads ~ 15C - 20C high. I went from the Albatron reading 15C - 20C low to the Abit reading 15C - 20C high. Why is it these guys cant get this right?

It undervolts a little.

BIOS = 1.575
IDLE
Speedfan: 1.55
CPUz: 1.536 - 1.552
Winbond HW DOC 1.55

Load
Speedfan:1.52
CPUz: 1.520 - 1.536 - 1.552
Winbond HW DOC 1.52 - 1.54

I haven't set up MBM5 yet.
 

Tom

Lifer
Oct 9, 1999
13,293
1
76
Thanks for the great info ! Have you compared hard drive performance ?

I'm curious if the lack of Intel Application Accelerator for non-sata setups makes any noticeable difference ?

 

oldfart

Lifer
Dec 2, 1999
10,207
0
0
Originally posted by: Dead Parrot Sketch
Thanks for the great info ! Have you compared hard drive performance ?

I'm curious if the lack of Intel Application Accelerator for non-sata setups makes any noticeable difference ?
I only checked it with HDTach. A better test would be boot up time, BF1942 map load time. You have to ignore the "min" readings. If you get one of those little spikes, it throws the min value way off.

845PE with IAA 2.3 WD 1200JB
HDTach 2.51
Max 49711
min 26170
ave 40550
burst 85.1
util 9.9%

875P with MS IDE drivers WD 1200JB
Max 49267
min 21540
ave 40531
burst 85.4
util 7.0%
 

jhites

Golden Member
Mar 19, 2000
1,854
0
0
Originally posted by: oldfart


I haven't really tried that hard to get more out of the CPU. 175 FSB on either board is not stable @ 1.6 Vcore. I wont go any higher. I haven't got down to trying 171 - 174 yet.

Lets see, Temp readings are off. Case temp seems pretty close, but CPU temp reads ~ 15C - 20C high. I went from the Albatron reading 15C - 20C low to the Abit reading 15C - 20C high. Why is it these guys cant get this right?

It undervolts a little.

BIOS = 1.575
IDLE
Speedfan: 1.55
CPUz: 1.536 - 1.552
Winbond HW DOC 1.55

Load
Speedfan:1.52
CPUz: 1.520 - 1.536 - 1.552
Winbond HW DOC 1.52 - 1.54

I haven't set up MBM5 yet.
Thanks for the compare. Nice work. :)
Did you ever get a chance to load MBM5?
Just curious what the low/high/avg was for idle and load on your board.

 

SinfulWeeper

Diamond Member
Sep 2, 2000
4,567
11
81
:cool: Nice work.
Not much of an increase from SC to DC though :(.

But I'll still build one in time. I need that DC for bragin rights :p.