Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Which is better or faster HX or TX?

378 views
Skip to first unread message

Technical Support

unread,
Feb 6, 1998, 3:00:00 AM2/6/98
to

Which chipset is faster HX or TX?

Steven Johnson

unread,
Feb 6, 1998, 3:00:00 AM2/6/98
to

Technical Support wrote in message <6bg0a4$q5f$1...@news.provide.net>...


>Which chipset is faster HX or TX?
>
>

Your ability to run both SDRAM and Ultra-DMA on the TX board both out
perform the HX capabilities,,,,,,,,,,I had a HX board,,,(TYAN TOMCAT
III),,,,it was great,,,,but for overall performance and stability,,,,,the TX
can not be beat,,,,,,,,
P.S. For another 6 months or so,,,,,(MOORES LAW)

Simon Karpen

unread,
Feb 7, 1998, 3:00:00 AM2/7/98
to

HX is faster, and unlike TX can cache more than 64MB. Its PCI bus
also seems to do better under heavy load (i.e. a usenet news server,
considering you're posting from teh support address for a news service)

--Simon

On Fri, 6 Feb 1998 17:50:23 -0500, Technical Support <te...@provide.net> wrote:
>Which chipset is faster HX or TX?
>
>


--
Simon Karpen s...@ntrnet.net
#include <std_disclaimer.h> My opinions are my own.
Proofreading services available: I will proofread any unsolicited
bulk/commercial e-mail for $5/word. To accept these terms, send me
unsolicited commercial e-mail. A bill will be sent promptly, along
with a corrected version of your advertisement.

Tan Wilson

unread,
Feb 7, 1998, 3:00:00 AM2/7/98
to

TX.

Technical Support wrote in message <6bg0a4$q5f$1...@news.provide.net>...

Rick Lindsay

unread,
Feb 7, 1998, 3:00:00 AM2/7/98
to

On 7 Feb 1998 03:34:21 GMT, Simon Karpen wrote:

:>HX is faster, and unlike TX can cache more than 64MB. Its PCI bus


:>also seems to do better under heavy load (i.e. a usenet news server,
:>considering you're posting from teh support address for a news service)

Under heavy graphic manipulation applicattions, within the 64mb cacheable
memory limitation of the 430TTX chipset, and when using SDRAM on the TX board
vs. MMX cpu and EDO on a 430HX motherboard, the 430HX is faster. This edge
becomes very significant if the 64mb limit of tthe 430TX chipset is exceeded.

SDRAM works on three cycles, EDO on two.

Two steps forward, three steps back, the Intel way....


Rick Lindsay, Lindsay Computer Systems, http://www.jumpnet.com/~lcs
Austin, Texas. 512-719-5257. Asus based systems, Asus Products.
Advanced Systems.
This message is SHAREWARE, please register...


KC

unread,
Feb 7, 1998, 3:00:00 AM2/7/98
to

HX
:)

Tan Wilson wrote in message <6bh7tf$us9$1...@columbine.singnet.com.sg>...

Dave Tatosian

unread,
Feb 7, 1998, 3:00:00 AM2/7/98
to

In article <6bh7tf$us9$1...@columbine.singnet.com.sg>,

"Tan Wilson" <wst...@singnet.com.sg> wrote:
>TX.
>
>Technical Support wrote in message <6bg0a4$q5f$1...@news.provide.net>...
>>Which chipset is faster HX or TX?

People don't run code on chipsets, they run code on systems!

HX makes for a better system!

JimP

unread,
Feb 7, 1998, 3:00:00 AM2/7/98
to

I have a computer with an HX and a computer with TX. I personally like to
use the one with the TX chipset. The HX chipset is a t2p4 ASUS board and
the TX is the ASUS TX97. Both boards have 64 mbs ram and are set up in
similar manor. The HX boots much faster. The TX has a 200MMX CPU. I have
not tried the 200MMX in the HX board yet but was planning on replacing the
166 some time soon.

JimP
From: JimP <nospam_...@hotmail.com>
Date: Friday, January 30, 1998 2:25 PM

From: James A. Pettyjohn <1jap...@hotmail.com>
Subject: spambot
Date: Friday, January 02, 1998 6:31 AM

Dear Spambot: Please make my elected officials
aware of your malicious activities:
pres...@whitehouse.gov
vice.pr...@whitehouse.gov
first...@whitehouse.gov
sen...@hutchison.senate.gov
sam....@mail.house.gov
Federal Communications Commission:
Chairman Reed Hundt: rhu...@fcc.gov
Commissioner James Quello: jqu...@fcc.gov
Commissioner Susan Ness: sn...@fcc.gov
Commissioner Rachelle Chong: rch...@fcc.gov

Fabian Robok

unread,
Feb 10, 1998, 3:00:00 AM2/10/98
to

Steven Johnson wrote:
>
> Technical Support wrote in message <6bg0a4$q5f$1...@news.provide.net>...
> >Which chipset is faster HX or TX?
> >
> >
> Your ability to run both SDRAM and Ultra-DMA on the TX board both out
> perform the HX capabilities,,,,,,,,,,I had a HX board,,,(TYAN TOMCAT
> III),,,,it was great,,,,but for overall performance and stability,,,,,the TX
> can not be beat,,,,,,,,
> P.S. For another 6 months or so,,,,,(MOORES LAW)

Looks like TX based boards are at least faster producing commas.
And that is probably their only advantage.

Fabian

--
------------------------------------------------------------------------
cand. ing. Fabian X. Robok _/ _/ _/ _/_/_/_/
Institut fuer Kommunikationsakustik _/ _/ _/ _/ _/
Ruhr-Universitaet Bochum _/ _/_/ _/_/_/_/
_/ _/ _/ _/ _/
Phone: +49 234 700 5388 _/ _/ _/ _/ _/
------------------------------------------------------------------------

Morgan J Tate

unread,
Feb 10, 1998, 3:00:00 AM2/10/98
to

Hi

I have an Abit IT5H HX ( revision 1.5 ) and had a Shuttle 569 TX based
system.
Having a choice on which to sell when I purchased my Asus P2L97 and 266 P2 I
decided to sell the Shuttle TX 569 board....Why....?

I now use Win 98 ( build 1650 ) so the question and endless frustration of
the TX chipset and bus mastering drivers is no longer an issue.

However having got an EIDE Yam 6x4x2 CD burner I was amazed at the 2mb
buffer almost being empty on the TX board and yet continually full on the HX
one. Using the same hard drive etc...... the only thing that was different
was the chipset on the motherboard. Also the hard drive was far faster using
the HX board than the TX one.
I tried many drives, Diamond Maxtors, Quantum Fireballs, Western Digitals
and Fujitsu's - mode 4 and UDMA drives, between the two systems and they all
seemed to perform better on the HX board.

I talked to a representative at Microsoft product support ( for W98 ) and
was informed that the drive performance, especially with the CDR issue, is
one that they are aware of with the TX chipset, and hopefully to resolve.

Having now bought a PL297 I will STILL suggest that the HX board is able,
either through the drivers and/or chipset, to perform better with EIDE
drives.

Not very scientific conclusion I know, not even quantified it with
benchmarks etc... But I had a choice and I believe that I chose the better
chipset.


Just look at the:
intel.other_components.pci_chipsets
and see how many satisfied owners there are with the TX chipset...........
I think not. !!!!!!

--
Regards

Morgan J Tate

please remove SPAM from email address.

morga...@hotmail.com

Please remove SPAM from Email address

Mike Sanzo

unread,
Feb 10, 1998, 3:00:00 AM2/10/98
to

Technical Support wrote:
>
> Which chipset is faster HX or TX?

TX. Sorta. It depends very heavily on the rest of the system. But
you're likely going to see TX boards with more capabilities than HX
boards, and, taking advantage of those capabilities, the TX -system-
will be faster. I can't remember the memory timings anymore, but I seem
to remember TX being better w/ SDRAM. <shrug> Chipset is hardly the
most important factor. I'd rather be concerned with additional features
(UDMA/33, SDRAM, etc), expandability (# of PCI slots, DIMM slots, etc),
reliability (that's the big one), etc...

-Mike

Eric Anderson

unread,
Feb 11, 1998, 3:00:00 AM2/11/98
to

Yeah, and speaking of speed, just wait until you want to install more than 64M
of RAM on a TX board. Kiss speed goodbye. The TX caching is crippled with
that 64M limit & I wouldn't recommend a board based on this chipset to anyone
when there are so many good alternatives available, like the SP97 for one. FIC
PA2007 for another.

-Eric

Dane

unread,
Feb 11, 1998, 3:00:00 AM2/11/98
to

Mike Sanzo wrote:
>
> Technical Support wrote:
> >
> > Which chipset is faster HX or TX?
>
> TX. Sorta. It depends very heavily on the rest of the
> system. But you're likely going to see TX boards with
> more capabilities than HX boards, and taking advantage
> of those capabilities, the TX system will be faster. I

> can't remember the memory timings anymore, but I seem
> to remember TX being better w/SDRAM. <shrug>

The TX-chipset can only cache 64MB of memory, so if you
plan on having more than 64MB of memory, stay away from
the TX-chipset. Get a good HX motherboard (that has the
2nd 'cache' tag ram, i.e., the Abit IT5H Rev 1.5 or 2.0

As for UDMA/33, although the HX chipset does not provide
suport for it, you can still use the drives. Make sure
that you run them in DMA mode 2 (they will be just as
fast). So UDMA/33 is not really a driving force.

As for SDRAM, the fastest timing possible with the TX
chipsets is 7-1-1-1 (10 clocks) and for 60ns EDO DRAM
it's 5-2-2-2 (11 clocks), so in theory SDRAM will be
slightly faster, however in practice it will only be
so when the L1 or L2 caches are flushed and refilled.

The penalty is closer to 1-3% by not using SDRAM. So
even this is not a big deal at all. However if you do
get SDRAM make sure you get the newer SDRAM with SPD
(it has an onboard EEPROM for timing information and
can be used with TX and LX chipsets).

> Chipset is hardly the most important factor.

Actually the "chipset" is the single most important
factor. There are many cloned chipsets available that
may have the necessary features but are very slow
(i.e., the fastest HX chipset is faster than many
cloned TX chipsets).

> I'd rather be concerned with additional features

> (UDMA/33, SDRAM, etc), expandability (#of PCI slots,


> DIMM slots, etc), reliability (that's the big one),
> etc... -Mike

Ok, so if you have at least a 200-MHz processor, you
can run your UDMA HDD in DMA mode 2 with an HX chip
set, so UDMA/33 support is not an issue. SDRAM while
slightly faster is not that big of deal especially
if you have more than 64MB in your system. Don't think
you'll ever need more than 64MB? Wait for this years
formal announcement of Windows98.

Windows consumes memory from the top down. So it will
start with the uncached memory above 64MB (between 5
and 30% slower). This more or less negates the 1-3%
speed advantage of SDRAM in TX chipsets.

If you are looking for good HX based motherboards,
checkout the Abit IT5H Rev 2.0. Priced from $104 to
$139 on the web. 4 ISA, 4 PCI (one shared ISA/PCI).
Soft menu which allows for easy modifications
(through the BIOS) of the clock multiplier, bus
frequency, and cpu voltages. 4 72-pin slots, 2 DIMM
slots. Great motherboard. Check the In-Win Q500 for
a good case (http://www.in-win.com/ino_q500.html) &
check http://www.megsinet.com/jcorwith/ultimate/index.html
for a good heatsink/fan combo (IERC P5455C2B, Sunon 60mm).

-Dane

--
Anti-Spamming measures. To reply to me, remove nospam.
and ROT-13 my email address: qn...@nospam.fbhguynaq.arg

Doc

unread,
Feb 12, 1998, 3:00:00 AM2/12/98
to

I generally agree with everything Dane wrote, BUT you should not overlook
the ASUS P/I-P55T2P4 which IMO still is the most overclocking-friendly MoBo
out there... It can be hard to find one now though, as they have been
discontinued...

Doc

Dane wrote in message <34E1F7...@nospam.fbhguynaq.arg>...

Steve Baron

unread,
Feb 12, 1998, 3:00:00 AM2/12/98
to

Micro X-Press had quite a few T2P4's last week. Rev 3.1 with 3.2

Doc wrote in message <34e23...@news4.image.dk>...

0 new messages