Whatever Happened to BTX?

J

Jasonx82

Supreme [H]ardness
2FA
Joined
Aug 14, 2004
Messages
6,706
I remember hearing about it in mid 2003. Now its middle of 05 Is BTX gonna replace ATX or BTX Dead? :confused: SOrry if this has been discussed i was just thinking and remembered btx :D
 
It's around, some people like Dell and Gateway are moving to it, which is suprising IMO. It does have more efficient cooling and all, but the transition costs outweigh the benifits in most manufacture's eyes. Especially with cool running AMD systems.
 
BTX faces even more problems than WTX did in becoming a standard: not only does it offer few to no benefits to most customers, support from its main creator (Intel) has all but fallen away now that the P4 (Netburst) line has been cancelled, and focus has been redirected to the P3 core (Pentium M), with as main features higher IPC and lower energy consumption.

Most likely we'll see BTX systems from the larger OEMs for a few years before this standard joins NLX, WTX and many others in the past.
 
I have a BTX systems and I love it. I keep putting off taking some pictures, but maybe i'll do that tonight. There are a couple of really cool things about BTX airflow.

BTX will be pimped big time when Pentium 4 Ds are in big supply. IMO, it will take a while for BTX to trickle down to DYI components.
 
pxc said:
I have a BTX systems and I love it. I keep putting off taking some pictures, but maybe i'll do that tonight. There are a couple of really cool things about BTX airflow.

BTX will be pimped big time when Pentium 4 Ds are in big supply. IMO, it will take a while for BTX to trickle down to DYI components.
Click to expand...
The P4 D is merely a stopgap solution for Intel to convince the world that they're not that far behind AMD. The P4 core was never designed for dual-core, however, and Intel is more than eager to replace the P4 D with a design which is not held together with duct-tape.

If you take a look at the dual-core Athlon 64, you'll see that it uses about as much power as previous single-core Athlon 64 CPUs.
 
Elledan said:
If you take a look at the dual-core Athlon 64, you'll see that it uses about as much power as previous single-core Athlon 64 CPUs.
Click to expand...
Huh? The Athlon 64 X2 CPUs have TDP ratings up to 110W. Single core 90nm Athlon 64s use much, much less power than that.

Duct tape. :rolleyes:
 
pxc said:
Huh? The Athlon 64 X2 CPUs have TDP ratings up to 110W. Single core 90nm Athlon 64s use much, much less power than that.
Click to expand...
I was mainly referring to the pre-Venice single core CPUs.

Duct tape. :rolleyes:
Click to expand...
What would you call it, then? You can hardly make the statement that Intel was planning to use the P4 for dual-core when it was first designed, like the Opteron and Athlon 64.
 
Elledan said:
I was mainly referring to the pre-Venice single core CPUs.
Click to expand...
Still doesn't matter. 90nm Athlon 64s run much cooler than the X2s, unless you pick extremes from each (hottest 90nm vs coolest X2) and then the difference shrinks to favor the 90nm single core by only 15-20%. Most other cases have the 90nm Athlon 64s up to 50% cooler vs X2 chips. Strangely, the Pentium D 820/830 has achieved TDP parity with the Athlon 64 X2 4200+, and the X2 4800+ only runs 10W cooler than the Pentium D 840.

Stoopid Guy above either didn't read my post or doesn't understand the FX55 and 4000+ TR used in that graph are 130nm versions.

Both the Athlon 64 and P4 have changes to allow dual core operation, but both use essentually the same method for dual core as for multi-processor systems. Like multiprocessor Athlon 64/Opteron systems, the bus that connects the cores is HyperTransport (point to point). The Pentium D, like multi-processor xeon systems, shares a front side bus (memory controller is in the MCH) for communication between the cores.
 
pxc said:
Still doesn't matter. 90nm Athlon 64s run much cooler than the X2s, unless you pick extremes from each (hottest 90nm vs coolest X2) and then the difference shrinks to favor the 90nm single core by only 15-20%. Most other cases have the 90nm Athlon 64s up to 50% cooler vs X2 chips. Strangely, the Pentium D 820/830 has achieved TDP parity with the Athlon 64 X2 4200+, and the X2 4800+ only runs 10W cooler than the Pentium D 840.
Click to expand...
What I meant was that in comparison with previous, pre-90 nm cores, Athlon 64 X2s do not use more power, meaning that current cooling solutions will be more than sufficient to cool those new dual-core CPUs.

I've also heard that Intel's TDP values are consistently rated lower than they really are. Whether this is true I don't know.
Both the Athlon 64 and P4 have changes to allow dual core operation, but both use essentually the same method for dual core as for multi-processor systems. Like multiprocessor Athlon 64/Opteron systems, the bus that connects the cores is HyperTransport (point to point). The Pentium D, like multi-processor xeon systems, shares a front side bus (memory controller is in the MCH) for communication between the cores.
Click to expand...
It's a bit more complex than that: http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=2419&p=2

Like I said, Intel's solution involves duct tape.
 
Elledan said:
Like I said, Intel's solution involves duct tape.
Click to expand...
I would say their inelegant design is more like building a truck by welding two cars together.
 
You must log in or register to reply here.
Back
Top