Trouble May be Ahead for Intel Medfield's Fight Against ARM

[Guest]

The comments on this site are horrendous. Come on guys! The article was actually a good one. It showcases some of the difficulties that Intel has ahead of them to break into this market. They know they don't quite have the right product yet. That is why they are busy in the labs completely redoing the ATOM for low power. Their next ATOM version will catch up with their desktop processes and that will be a product they can compete with. Until then, they can't sit around and wait. ARM is attacking the sector aggressively and they must put something out now. A 32nm SOC based medfield is the best they can do and they can make it an interesting product. In won't quite compare in power savings, but it can in performance. They also have the Intel manufacturing muscle working for it. They have to make some inroads so that 2013 can be the year they hope for (a year they want to dominate with an actual product that can beat ARM).

ARM is in the same dillema really. They are coming from the mobile space and now want to join the laptop sector. Problem is, much like Intel, they don't have a compelling product. Their products save power but lack greatly in raw performance. Laptop users won't take kindly to that. The lack of compatibility won't make them happy either. Basically, both will put out subpar products in 2012 in hopes of gaining a foothold for a much bigger race in 2013 and 2014. 2012 will be interesting, but wake me up in 2014 when these two are going head to head in just about every sector.

 

[Guest]

Didn't I see that Intel demonstrated a prototype Medfield tablet that had a terrific and oh so cool innovation....... a fan to make it even cooler! Wowee.

 

[Guest]

zingam: agreed mate, you've nailed it on all fronts.

Before anybody goes around shouting that Windows 8 will be a resounding victory for ARM, Microsoft, or anybody else, they ought to go try the Win8 developer preview. Once the shock of what a turd it is starts to wear off, you can then return to the fact that tablets are still an inferior way to run desktop apps, and if you add a keyboard, well, that tablet is just a netbook then, just slower and smaller than a laptop.

 

[shqtth]

Windows tablets are good. Converted my EEE to one. There is plenty of touch launcher programs for windows tablets making loading apps easy (asus makes one for their touchscreen eee tops). Doesn't matter if an app is made for a tablet or not. Its made for a mouse. So already point and click will work. multitouch is another story depends on the app, but either way, all apps work on the tablet. And I am sure you can use 3rd party or 1st party widgets for text input or directional inputer such as virtual keyboard or virtual game pad. 😉

I hacked my xp to support features of the tablet edition and features of vista 32bit for tablet text input. Worked really well.

 

[Guest]

"Remember this: Intel has failed every single attempt to design a product and enter a market where they are not the only dominant force. (and are not able to use anti-competitive practices to beat their competition) Maybe it is their marketing or maybe they are not that great engineering company after all.
Well one thing they do well is their production process. At that they are really the best."

You mean like the server market. You mean like the ssd market. You mean like the core logic market. You mean like the RISC market (i860). You mean like the mobile market (xScale). Please spare us the retarded comments about things you have absolutely no clue about. Intel never entered the video card market. It had to do with a lot of factors... one being the move of the market to smaller form factors and the fact that there was much more potential for growth using an on chip product. Larrabee has thus become an add in card that resembled more of a stream processor with real CPU capabilities for server products. A smart move that will keep the competition at bay and keep Intel in the high performance computing game for some time. I guess you can call Itanium a failure except that they have sold a lot of those things in a market where volume is extremely low. They have used Xeon to cover every other sector of the high end server market and the latest Xeon's have much of the same mission critical technology as Itanium.

Basically, if Intel is not a success at entering markets, then I don't know what the definition of success is. They have entered a lot of markets and had little resistance taking over. We hardly hear of companies like SGI anymore and Digital Equipment sold much of their tech to Intel (oh where for art thou VAX). SUN sells mostly Intel systems now as does IBM and HP server division. Cray doesn't build MIPS systems anymore. Damn near everything in your system is Intel Inside and they even partner up with memory maker Micron. They have championed USB, PCI, SATA, and now LightPeak. They have pushed WIFI standards and lots of little standards that you don't always see. Please try reality next time you post.

 

[Guest]

IVBMan: That's some bizarre logic you have. Servers are core logic are just an extension of desktop CPUs, I'm not sure how on earth you came to the conclusion that Intel didn't have a head start there with their x86 monopoly. Then you go on to trumpet off all of their failures off as if they were successes. I think the Intel i860 wikipedia page says it best:

"...Andy Grove BLAMED the i860's FAILURE in the marketplace on Intel being stretched too thin..."

Intel doesn't dominate SSDs, and if anything, they have an awful reputation for reliability, their SSD firmware is about as bad as their GPU drivers. I don't have exact figures for how much money Intel makes on SSDs, but I doubt it's a significant revenue stream for them, it falls in the category of "don't quit your day-job".

 

[Tamz_msc]

The most important factor will be the pricing. Intel will be able to pull it off in terms of performance and power consumption but they must get their pricing right.

 

[Guest]

Medfield , another Larrabee in the making

 

[PhilFrisbie]

Intel made a big mistake when it sold its XScale ARM business. I still have a Toshiba e740 PDA with a 400 MHz Intel StrongARM/XScale CPU. That PDA still works fine, although hampered by running Windows CE.

 

[kronos_cornelius]

"do-or-die" ? isn't that a little hyperbolic ?

Intel has at least 5 years to enter the mobile arena. It will be do-or-die once we start seen ARM based servers and desktops. And that is still a few years away.

 

[eddieroolz]

I wouldn't say any architecture is superior over others. It's like people saying a hexacore is is better than anything else for all applications. You'd have to be a fool to make that declaration these days, especially after benchmarks that show hexacores excel in its own applications and such.

x86 has its uses. ARM has its uses. All I care to see, is if Intel can get the heat and power usage of x86 mobile processor down to a manageable level.