- Thursday, 9 July 2009
- INQ Mobile
- RSS
Although prepared for martyrdom, I preferred that it be postponed - Winston Churchill
X48 is a horrible idea
SIGH. DEEP SIGH. No sooner does AMD buy more shells for its shotgun podiatry vaudeville act than Intel returns to its old ways.
That is not, for those new to this game, a compliment.
The problem is the X48 chipset, quite possibly the second best chipset not on the market yet.
It is a minor, and I mean minor revision to the otherwise excellent X38 chipset. The added 10 merely brings you from a 1333FSB to 1600FSB. It is otherwise the same part, one look at the mobos that have the X38 sticker hastily crossed out and the name X48 written in will tell you that.
The problem? In the days of yore, like six months ago for people with memories that long, this would not have rated a name change, or anything else, just a bump in the stepping noted on the technical documents. New boards would have a "now with more vitamins, minerals and FSB" sticker on it, and the world would go on.
What do we end up with? A "new" chipset on "new" boards, and ghod help us, fodder for the marketing critters. All this for functionality that was either there from day 1 on X38, or added in the last month, then almost assuredly fused off for 'new vanilla' X38.1 parts not labeled X48.
Why did they bother? Is it to milk money out of the high end? Don't they do enough of that already with the EE/XE/whatever chips? Are they trying to make us buy an XE mobo for your XE chip, and woe betide the fool who buys one without the other? Are they going to EOL the X38 after a whopping 39 day lifespan?
In the end, we are left with two chips that do exactly the same thing with one bullet point artificially changed. OEMs now must stock multiple SKUs, must have multiple variants of the same mobo, and most likely pay more for the added 266MHz. What a mess.
Who pays? In the end, the customer - that would be you and I. For anyone touching this mess, from Intel to little Timmy buying the uber-rig, it only adds headaches, which in turn jack up costs. There was a time, again in the golden days of six plus months ago, when Intel would not do such a thing because it realised it shafts the very consumers who give them the highest margins. Oh how I long for those days.
With any luck, Intel will realize that this course of action is not a very bright one, and will backpedal ASAP. If it wants to go back to the bad old monopoly days and jack up prices, fine, it earned it, but just stop playing SKU games that hurt everyone. µ
Share this:
Business Intelligence Software
Customer Relationship Management (CRM)
Enterprise Accounting Software
Enterprise Resource Planning (ERP)
Enterprise Systems Management
The excellent i865 and i875 chipsets had only the PAT feature added to the latter as a real difference and it was found out that i865 had PAT disabled by default but many manufacturers enabled it regardless of Intel will. So what ? 
It's just like nowadays manufacturers are allowing FSB1600 on X38 motherboard although it's not certified above FSB1333 functionality. 
The customer deserves to be shafted by Intel. If you're dumb enough to contribute towards a big monopoly, what the hell do you think is gonna happen!?!!
the customer does not prevent monopolies, he will endure them. A company will only be competitive, if it has competition. And right now, I'm enduring Windows, and sorry I am for it.
I expect the x48 yield 2000 FSB out of the box, for each and every board. That's what the higher bin presumably means in this case.

The official Intel support for FSB 1600@x48 sounds retarded, since Asus e.g. touts their guaranteed support for rated FSB 1600 in each and every x38 board. 

So unless ALL x48 boards go 500MHz with the MHC, we have the utter most biggest sham release in the history of motherboard chipset.

And so, maybe we'll get a 500fsb QX9X50 guarantee, who knows...? :D
So let me get this straight. The customer should support the inferior product with hard earned dollars in order to prevent a better company from becoming a monopoly? Since when is that sound logic? It's not the job of the customer to break up monopolies, the customer will support the better product, period.
X48 is said to be basically a 'cherry picked' X38 chipset resulting in better overclockability and official support for a 1600MHz FSB. 

The "EIST bug" present in existing reference X38 boards which X48 fixes.

The gigabyte X38 chipset motherboard GA-X38-DQ6 has 1600FSB see http://www.giga-byte.com/Products/Motherboard/Products_Overview.aspx?ProductID=2665

The x48 is just a rip off by intel - Chipzilla fat cats, a clear marketing rip off.
I hope you're the first to be assimilated, you spineless consumer twonk.
BOOOOOOOOOOO!!!
@windows monopoly: Could be far worse. Could be an Apple monopoly. If that were the case we wouldn't even be considering the chipsets that our motherboard has as they'd be picked by Apple for us.

I don't see how buying the faster chip is the stupid decision. If you want to pour money into the over priced, under performing Phenom coffers I guess that's your choice.
i925XE, i955, i975X anyone?
Thanks for reminding us that intel has'nt really changed forever, just until they reclaimed market share and could coax us all back onto their gravy train.
Why don't you complain about something valid like how the latest and greatest QX9650 only works on Intel chipsets.
Charlie, can you give some info about the X38 chipset "EIST bug" ? No matter I try I can not find concrete information, there is just an endless repeating of "EIST bug" but nobody know what it is. Thanks..
THAT WAS IT, MEDIA GETTING GOOD AS 333 BECOMES NOR M. Its' too bad best is Last. It took forever to Dream up this.

I am Glad XP has higher Ground Now. I am anxiously awaiting first ULTIE Shots. It seems getting things wrong till gotten right are Norm. Like Charts of two contollers of in/out back/forth 2 completely discrete lines crosing back & forth, makes no difference as On Crossbar, Just little More V. Is where naughty 790 is heading INTO x2 CARDS, so maybe better design in concept, Ultie Starting Down Road From There?, awaiting its perfection w/ultie.. Take That, INTEL. 

COME ON ULTIE, COME ON 4500, COME ON 100.come on amd_Ultie64. Au64.Hey U .wake u p.
thomas vondrashek
I really don't see such a problem here. If they add what they consider a new feature, or higher performance, they can call it whatever they want.

If they feel it's worth more money, so be it. 

No one is required to buy it.

Stop whining.
"The customer should support the inferior product with hard earned dollars in order to prevent a better company from becoming a monopoly?"

If you say that Intel has better business tactics, yes, I would agree with you.

Technologically, Intel's still playing catch-up with the integrated memory controller, and ditching the age old FSB so that multi-socket can be better implemented.

I sound like a horrible AMD fanboy but frankly the few millisecond to seconds of difference between what Intel and AMD offers is not enough for me to justify paying the Intel bully. Their integrated graphics chips are crap; they're copying things that their rivals have done years ago. The only reason why they're "good" is that they're big, has a R&D budget like you wouldn't believe, and have enough cash to run insane PR programs.
... DDR3 wasn't ready, the chipset had a bug, etc.

The pity about 925 and 955 was that there were never any good CPU's for them...
Option A: Use less marketing "techniques" (or "tricks", depending upon who is doing the whinging) and obtain less profit

or

Option B: Use more of the same to obtain more profit

Hm. Gee. TOUGH to figure out, this one.
There were more differences between the i865 and i875 chipsets than just PAT. If nothing else, i875 supported ECC memory. Funny thing about the X38 (and probably X48): apparently ECC is supported for DDR, but not for DDR3.
I don't see how this product naming change of the same chip with only a bullet point missing is any different from what has been going on in CPUs... for AGES AND AGES.

It's just binning, or artificial marketing binning. You've never complained about it in CPUs where it has blatantly been going on, so why complain about it now. You keep talking about the costs of having new stickers with "X48" written on them. Does the 4 really use that much more ink than the 3? Are stickers that much more expensive for motherboard vendors than for CPU resellers?
In my book X48 rates like a 8800 Ultra. No difference from the previous board except for slightly higher clocks, cherry-picked GPU's and a price tag that says "This is how stupid we think you really are."

Nice going, Intel. We're just not buying it.
Also, I hear they are shorting a pin to disable DDR2. Its all fun in the sun baby, DDR3 is a complete rip off, has crap latency, and has no ECC easily available.

Intel support, which used to be one of the best, is now a bunch of pot smoking retards in Oregon that lie and know less about the product than the consumers.