Athlon64 3400+ Socket 939 benchmarked in Sandra

***Veridian3*** · Mar 22, 2004, 01:30 PM

Computerbase.de have posted benchmarks comparing Athlon 64 S754 and S939 chips, you can check them out by clicking the link below...

Full Benches

**Roadee** · Mar 22, 2004, 01:41 PM

It is only a 512kb cache chip though.........

Zardon · Mar 22, 2004, 01:43 PM

Quote:

Originally posted by Roadee
It is only a 512kb cache chip though.........

El Kapitano · Mar 22, 2004, 02:33 PM

The extra 185 pins seem to cause some extra drag.

Is it dual channel? I don't see any memory bandwidth tests on there..

Vampyromaniac · Mar 22, 2004, 09:06 PM

I read Anandtech some time ago, the socket 939 chips will be as follows:

Athlon64...................Athlon64-FX
--------------------------------------------
dual-channel..............dual-channel
512K cache...............1024K cache

whereas the current chips are:

Athlon64-754.............Athlon64-FX-940
---------------------------------------------------
single-channel.............dual-channel
1024K cache..............1024K cache

the exception being the Newcastle 3000+,
which as I understand the 939 regular64's will be using as well (and so the cache is the same 512K)
until they change to 90nm

mkk · Mar 22, 2004, 09:31 PM

Quote:

Originally posted by El Kapitano
I don't see any memory bandwidth tests on there..

Wich is especially notable since the only part of the Sandra benchmarks that is just maybe useful is the memory bandwidth.

512k L2 cache does seem to become the norm for Socket 939 Athlon64's being released in the near future. It should help production without any real loss of performance in most applications.

@/2ct!< · Mar 23, 2004, 01:52 AM

But this in no way seems to justify the PR rating of 3500+ for the 939 one...

Vampyromaniac · Mar 23, 2004, 06:45 AM

Well, they did have it running in single-shannel in the tests. And you must admit, the tests they used are by no means a comprehensive evaluation of the chip

let's just wait until they are out, and some real reviews are made.

@/2ct!< · Mar 23, 2004, 11:02 PM

Yes, I said this does not, others may, but sometimes dual channel didnt seem to help very many athlons b4

Monkeyboy · Mar 25, 2004, 10:45 AM

I think the difference in benchies between the 754 and 939 versions are purely down to the chipsets used - 754 used a Via K8T chipset, while 939 used a nForce 250.

Mar 22, 2004, 01:30 PM	#1
*Veridian3* Administrator Join Date: Nov 2002 Location: Cloaked Posts: 5,028 Rep Power: 182	Athlon64 3400+ Socket 939 benchmarked in Sandra Computerbase.de have posted benchmarks comparing Athlon 64 S754 and S939 chips, you can check them out by clicking the link below... Full Benches

Mar 22, 2004, 01:41 PM	#2
Roadee Never forgotten Join Date: Aug 2002 Location: Rest In peace, Joe. Posts: 2,198 Rep Power: 69	It is only a 512kb cache chip though......... __________________ We are born naked, wet, and hungry. Then things get worse.

Mar 22, 2004, 09:06 PM	#5
Vampyromaniac confutatis maledictis Join Date: May 2002 Location: somewhere dark Posts: 5,974 Rep Power: 77 System Specs	I read Anandtech some time ago, the socket 939 chips will be as follows: Athlon64...................Athlon64-FX -------------------------------------------- dual-channel..............dual-channel 512K cache...............1024K cache whereas the current chips are: Athlon64-754.............Athlon64-FX-940 --------------------------------------------------- single-channel.............dual-channel 1024K cache..............1024K cache the exception being the Newcastle 3000+, which as I understand the 939 regular64's will be using as well (and so the cache is the same 512K) until they change to 90nm __________________ *Digitalis 3.3* Athlon 64 3000 // ASUS K8V SE Deluxe // 1024MB PC3200 (2-2-2-10 1T) ATI All-In-Wonder 9700 Pro // 20" Dell 2005FPW (DVI) M-Audio Revo 7.1 + Philips Acoustic Edge // Klipsch ProMedia 2.1 320/16 Western Digital WD3200KS + 120/8 Seagate 7200.7 NEC ND-3550A 16x DVD±RW + Lite-On 52x24x CD-RW Antec Sonata case // 480W Antec TruePower personal bests \|\| Aq'3: 46796 \| 3D'01: 20461 \| 3D'03: 6336 \| 3D'05: 2677 \| PC'04: 4605 \| PC'02: 7691,9092,1250

Mar 23, 2004, 01:52 AM	#7
@/2ct!< HardwareHeaven Senior Member Join Date: Aug 2003 Location: Not Vegas, anymore :( Posts: 1,391 Rep Power: 0	But this in no way seems to justify the PR rating of 3500+ for the 939 one... __________________

Mar 23, 2004, 06:45 AM	#8
Vampyromaniac confutatis maledictis Join Date: May 2002 Location: somewhere dark Posts: 5,974 Rep Power: 77 System Specs	@/2ct!< Well, they did have it running in single-shannel in the tests. And you must admit, the tests they used are by no means a comprehensive evaluation of the chip let's just wait until they are out, and some real reviews are made. __________________ *Digitalis 3.3* Athlon 64 3000 // ASUS K8V SE Deluxe // 1024MB PC3200 (2-2-2-10 1T) ATI All-In-Wonder 9700 Pro // 20" Dell 2005FPW (DVI) M-Audio Revo 7.1 + Philips Acoustic Edge // Klipsch ProMedia 2.1 320/16 Western Digital WD3200KS + 120/8 Seagate 7200.7 NEC ND-3550A 16x DVD±RW + Lite-On 52x24x CD-RW Antec Sonata case // 480W Antec TruePower personal bests \|\| Aq'3: 46796 \| 3D'01: 20461 \| 3D'03: 6336 \| 3D'05: 2677 \| PC'04: 4605 \| PC'02: 7691,9092,1250

Mar 22, 2004, 02:33 PM	#4
El Kapitano Most of the 7 dwarfs in 1 Join Date: Jul 2002 Location: Rochdale, Lancashire Posts: 947 Rep Power: 0	The extra 185 pins seem to cause some extra drag. Is it dual channel? I don't see any memory bandwidth tests on there..

Mar 23, 2004, 11:02 PM	#9
@/2ct!< HardwareHeaven Senior Member Join Date: Aug 2003 Location: Not Vegas, anymore :( Posts: 1,391 Rep Power: 0	Yes, I said this does not, others may, but sometimes dual channel didnt seem to help very many athlons b4 __________________

Mar 25, 2004, 10:45 AM	#10
Monkeyboy DriverHeaven Lover Join Date: Aug 2002 Location: Vista: the woe starts now! Posts: 113 Rep Power: 0	I think the difference in benchies between the 754 and 939 versions are purely down to the chipsets used - 754 used a Via K8T chipset, while 939 used a nForce 250.