nav search
Data Center Software Security Transformation DevOps Business Personal Tech Science Emergent Tech Bootnotes BOFH

back to article
Intel is upset that Qualcomm is treating it like Intel treated AMD for years and years

The old addage" Reap as you sow"

33
1
Silver badge

Qualcom

..and the Qualcom chips are arguably better too..

19
2

Re: Qualcom

I've no strong preference, although I've certainly no arguments that Qualcomm make great soc's. I do, however, back them in this spat because Intel have happily shown shitty behaviours in the past and having a monopoly on all platforms is even worse than on a single one.

21
1
Silver badge

Re: Qualcom

Irrelevant, though as Intel seems to have lost mojo (Atom embedded bricking, Cable Modem chips that don't work properly, killing off Atoms and wearables, Itanium, 64 bit Atoms that can ONLY have 2G RAM), probably true.

9
0
Silver badge

Re: Qualcom..and the Qualcom chips are arguably better too..

Which is presumably what Intel is really upset about.

17
0

Re: Qualcom

Your suggesting because Intel was a bad player we should now support another bad player?

I think Intel was prosecuted for it's behavior, and I think Qualcomm should be also. If I am wrong about Intel then they should both be prosecuted.

5
4
Bronze badge

Uh, *raises hand*

Qualcomm filed with the ITC to block iPhones that use Intel modems after Apple has decided it and it's phone-makers won't pay any royalties to Qualcomm. Intel can spin it anyway they want because... Who else uses Intel modems?

So, if Xolo makes the X900 handset (2012) with Intel inside, but they pay Qualcomm the contractual royalties for SEP patents... Might find that no one is dumb enough to trigger anti-trust when your IPR is probably just as valuable as winning the socket.

AMD needed to copy the Intel architecture or you have something that won't run the same software.

Isn't it obvious Qualcomm has technology that has nothing to do with the air interface standard (SEP patents) and since Apple is famous for being difficult to work with, they opt to not license these critical chip patents to Apple?

Remember these are not SEP patents.

Can Apple make a System-on-a-Chip(including the modem) and compete on cost with "landfill android" handsets?

Well?

3
15
Silver badge

Re: Uh, *raises hand*

AMD needed to copy the Intel architecture or you have something that won't run the same software.

Hmm, well I think it was more to do with Intel being obliged to reach a cross licensing deal with AMD to avoid monopoly investigations reaching damaging conclusions.

And of course Intel's chips are these days copies of AMD's ISA; it was AMD who developed the 64 bit architecture we all use today. Intel decided to go 64bit with Itanium; we all know how badly that went. AMD did something different, which worked, and Intel had to copy it.

In fact Intel copied it so early on that their first copies were "emt64", which lacked some of the instructions that amd64 had when it first hit the market. There's still differences today, but compilers generally work around this problem without difficulties.

41
2

Re: Uh, *raises hand*

More nonsense.

Intel had x86-64 prototypes in labs way before AMD.

2
20

Re: Uh, *raises hand*

That's nice for their labs. AMD was first to market with something usable. If Intel had their way those chips would probably still be in the lab.

16
1

Re: Uh, *raises hand*

> AMD who developed the 64 bit architecture we all use today...

Check your history....as

DEC developed the 64 bit architecture we all use today....1988-1990 time frame...

a large part of development engineers went to AMD from DEC before Alpha went to Intel

as result of a lawsuit by DEC vs Intel.....

https://en.wikipedia.org/wiki/Alpha_21064

https://en.wikipedia.org/wiki/DEC_Alpha

http://alasir.com/articles/alpha_history/dec_collapse.shtml

1
10
Silver badge

Re: Uh, *raises hand*

"AMD was first to market with something usable."

More than just "usable".

We bought one of the first ever AMD64 laptops - 1400 by 1050 screen. With an uprated hard drive, it made all the desktops look quite sick and was a useful development platform for 5 years.

I remember interviewing someone somewhat later who told me rather patronisingly that 64 bits isn't necessarily faster than 32. True, I said, and then showed him an Asus 64 bit gaming laptop running a database query with a 14Gbyte process.

But for AMD, I suspect we would still need big tin for that sort of thing.

3
1

Re: Uh, *raises hand*

So, DEC developed x86-64? No.

Nobody is saying "AMD was the first to develop a 64bit architecture", what they are saying is that it was AMD that developed the 64bit extensions to the x86 architecture.

In fact, there was plenty of suggestion at the time that Intel wanted to use a different (and incompatible) set of their own x86-64 instructions, but were shot down by MS who were already developing Windows XP for AMD64 - (as well as Itanium at the time, plus x86), and told Intel they would not develop for yet another instruction set. That's what El Reg suggested here at the time: http://www.theregister.co.uk/2004/01/29/intel_likely_to_offer_64bit/

5
0
G2
Coat

.

Pot, meet kettle.

23
1
Gold badge
Unhappy

Demonstrates the default "business personality" is still sociopath

And as we know sociopaths will do whatever they can get away with.

"It's just business" as Don Corleone pointed out.

Film goers will recall what happens to people who fail to show the proper respect to the Don.

15
0
Silver badge
Mushroom

How did we get here?

" ...is Qualcomm's only remaining competitor in the merchant market for premium LTE baseband processor modems "

Forget a moment about relative quality, or Intel's dubious practices. Qualcomm lobbied to have an inferior 3G (too much based on CDMA because of their patents) bought companies such as Flarion and buried them purely to try and control all the 4G IP. The LTE is an international standard and patents to do with it should be under FRAND. There should be no "double dipping", Qualcomm essentially making chip, not as the business, but as a means to control the royalties and charge a percentage on the complete product. That should be illegal everywhere.

Qualcomm only make chips to strengthen their main activity, Patent Trolls.

16
6
Bronze badge

Re: How did we get here?

If you are the subject expert on the standard, as Intel is for desktop CPUs, who in their right mind would put all that development solely in the hands of lesser companies to develop the chips?

Qualcomm had to enter the handset business in the mid 90's as, despite considerable hand holding to Motorola, Sony, and Samsung, no one else could design a carrier-accepted handset that delivered on their promise of excellent coverage and call capacity.

Qualcomm won the socket for the iPhone4s after the huge embarrassment of the "death grip" problem with the Infineon/AT&T GSM/UMTS-only iPhone4. Qualcomm still makes the only chipset in the world that worked on all bands and providers. That "world roaming" thing.

"Qualcomm lobbied to have an inferior 3G"

Or, Qualcomm's 3G expertise and technical assistance to networks around the world is the only reason UMTS wasn't murdered in its sleep.

Make UMTS suck less, you sell more handsets.

You collect royalties? Well, there you go.

What do you call an LTE-only phone? A really expensive and shiny flashlight.

5
8
Bronze badge

Re: How did we get here?

"The LTE is an international standard and patents to do with it should be under FRAND. "

You don't get it.

Clearly, there is a patented radio design that Qualcomm leverages to offer the only acceptable internal "world roaming" modem solution and they have chosen to not license it to others.

Everyone has these kinds of unlicensed patents, even Apple and Intel. Do you understand this?

"world roaming" in a really tiny space will never be considered "Standard Essential".

Qualcomm has critical chip-design patents that Apple needs that are, genuinely, not SEP.

End of story.

5
7

Re: How did we get here?

That doesn't describe the patents Qualcomm sued with. Name the ones you're talking about or shut up.

1
0
Bronze badge

Re: How did we get here?

"Name the ones you're talking about or shut up."

With pleasure.

To answer your question, you need to know how radios have been working since the 50's. Start with the building blocks.

There are 28 Qualcomm patents covering the supposedly simple voltage controlled oscillator.

26 patents that cover RF mixers

Do a boolean search on USPTO with the words synthesizer and Applicant Name: Qualcomm.

Also, do a boolean search with the normally innocuous word "digital" and Qualcomm. About half of those are important, assuming they were the solution chosen for the final product.

Since Electronics is not your field, 1G to 5G uses superhetrodyne downconversion because nothing else has been proven to work outside the lab. A simple explanation is you hang an antenna off of a RF amplifier, filter out the crap you don't need, mix it against a local oscillator and BLAM! FM radio.

For instance, Qualcomm released a Zero-IF receiver design about 15 years ago. ICERA didn't lose market because they couldn't make a modem or they couldn't get a license. They didn't invent their own Zero-IF radio architecture. Exhibit A? Surface 2 with the AT&T 4G modem.

All of these blocks in a superhet receiver have been around since the beginning of FM radio.

The problem for you is nothing, absolutely nothing I've discussed can be construed as SEP.

That's just receivers. I haven't started with transmission except VCO's, PA's, synthesizers, and all sorts of crap are also critical but since they are applicable to all RF standards to some degree, can in no way be construed as SEP.

2
2
Silver badge

Re: How did we get here?

"A simple explanation is you hang an antenna off of a RF amplifier, filter out the crap you don't need, mix it against a local oscillator and BLAM! FM radio."

Actually superheterodyne receivers were originally for AM; you beat the incoming RF with a higher frequency to get a low frequency output which is then amplified through a fixed frequency tuned amplifier. Superheterodyne was used because AM signals were mostly quite low frequency - down to about 150kHz - so the mixing frequency needed to be higher not lower. Because FM needs more bandwidth than AM, the intermediate frequency on FM tended to be much higher - around 10MHz rather than the 455kHz used for AM - but the principle is exactly the same.

Also, superheterodyne downconversion is as redundant as PIN number.

Making it work in the 5GHz band though, fills me with awe.

4
0

Re: How did we get here?

You have still not named the patents Qualcom has sued with or how they make your point, thus please continue to shut up.

1
0
Flame

Re: How did we get here?

Only because they have used their patent library fees to squeeze out the competition.

I know because I worked for a UK company that got killed by just that.

0
0
Anonymous Coward

Re: How did we get here?

@whoseyourdaddy - I worked at Icera after acquisition and we had a good product, got into some phones but there was always a patent library in the background.

Best job I ever had though, a change to do great work, great people and no regrets.

0
0
Silver badge

There are a lot of similaritites.

There are differences however.

Intel screwed AMD which had WORKING and often SUPERIOR gear. In the mid-2000-es I would have taken an Athlon instead of a Pentium 4 any day. I would still take an AMD A series instead of an Intel Core for consumer use any day too. The graphics performance is light years ahead of an Intel IGP. Always was and still is. The IT in my previous job, tried to mandate I use a Core 7 laptop with an Intel IGP card instead of my old A4 machine. It tried it. For 15 minutes. You could see the windows updated on a 4K monitor. It was like using Windows 95 with an ISA VGA card. While my reaction was "screw that", I have to admit - it worked.

That is NOT the case in mobile, CPE, etc. Whatever we say about Qualcomm, while it may have an "interesting" (quotes intended and needed) business model, its gear actually works. Ditto for Broadcom. Now compare it to the lousy runts Intel prints for LTE and Cable chipsets. Neither can be considered fully functional and working.

26
3
Bronze badge

In the end, Android wins...

I suspect this is what Tim cook is really pissed about:

https://www.qualcomm.com/news/releases/2013/02/21/qualcomm-rf360-front-end-solution-enables-single-global-lte-design-next

No SOC for you....

6
2

lose lose

unless the courts give both a slap and allow at least one third party into the scene.

1
1
Bronze badge

Re: lose lose

"unless the courts give both a slap and allow at least one third party into the scene."

Like Intel?

I actually believe Apple can outdesign Intel/Infineon based on the strength of their app processors alone.

But, outdesign Qualcomm's radio using their own IPR, SEP IPR, and IPR they can *license* from others?

I think this struggle is what caused the problem in the first place.

4
0
WTF?

I had to laugh at Intel's sheer cheek here. With a story like that, I hope their lawyers have considered the penalty for perjury!

Apple might well design their own, but they'd have to hire the relevant hardware experts because its not so easy to do right, just ask Intel. Mind you, Intel have problems trying to make pretty much anything work properly, they have launched so much dud hardware they are (in)famous for it.

8
1
Anonymous Coward

confused

Qualcomm and Apple are in dispute, and as part of that dispute Apple has ordered partners not to pay Qualcomm. Qualcomm responds by requesting a ban on Apple products that (actually or allegedly) use the IP that's part of the dispute & for which it should be paid.

Intel is not a party to that dispute, but is potentially hurt by it. So, rather than talking to Apple about the whole thing being settled and the problem going away, their lawyers go after Qualcomm? This is the same Intel that basically lifted the AMD64 instruction set, is not unknown to use lawyers rather than innovation as a "source" of "competitive edge", & now going after a competitor because of a third party dispute, and doing so in a way that just happens to have the potential to favour Intel in the marketplace?

What happens if Qualcomm says "Yes, Your Honor, we acknowledge Intel might be hurt by this, and we are very sorry. Perhaps, rather than using lots of the court's time, they could actually help resolve it by talking to their friends at Apple?"

4
2
Bronze badge

Re: confused

"Intel is not a party to that dispute, but is potentially hurt by it."

Eh, does anyone else use Intel modems besides Apple?

The problem boils down to is if you buy Qualcomm chips and have a royalty bearing license to sell phones with those chips, Qualcomm licenses are "flat rate" as those who use Mediatek or Icera modems are not affected.

But, if you don't buy Qualcomm chips and flatly refuse to pay any royalties, at all.

Well... Intel needs to choose better customers.

1
0

More nonsense "Oh poor AMD ... Intel is so evil" crap!

Enough is enough!

The whole 'net is filled with AMD shills bashing Intel and Nvidia spreading lies and telling the world that everyone must love the Ryzen and Vega crap. AMD products are bad and unreliable.

AMD cheats and spreads false info and fake benchmarks. Its products are a huge mess and a waste of money.

Now they spread a lot of lies against Intel claiming that all of a sudden Intel CPUs TDP would be a lot higher and other nonsense like that.

At the same time they are releasing the Vega GPU with a 375Watt claimed TDP ... which means 600Watt+ for real. Because AMD never tells the truth. When they claim that their Ryzen CPUs would be in the 130Watt TDP range would mean 250Watt+

The most disgusting thing is al the online sites promoting the AMD lies and bashing Intel and Nvidia instead.

This has to stop.

AMD deserves to go bankrupt. Is a disease in the IT industry.

2
21

Re: More nonsense "Oh poor AMD ... Intel is so evil" crap!

Just wondering, do you have to go all the way to Santa Clara to collect your cheque from Intel for that, or do they mail it through to you?

9
2

Re: More nonsense "Oh poor AMD ... Intel is so evil" crap!

I do not work for Intel or Nvidia. BUT you AMD shills surely are getting a fat check for all the lies you are writing on the 'net bashing Intel and Nvidia products while promoting the low quality AMD ones.

All the noise you are making on the 'net is just plain disgusting. You invaded all forums and blogs and comments sections all over the 'net. AMD must be paying a lot of money for the illegal bashing campaign. Worse the products are more the scam companies use illegal campaigns to bash competitors and promote their crap.

1
15
Gimp

Re: More nonsense "Oh poor AMD ... Intel is so evil" crap!

No one mentioned nVidia. Intel are big enough and ugly enough to fight their own battles and have frequently been caught behaving badly (hence the comments regarding past episodes with AMD in the x86 arena) so they probably don't need your attack dog like assistance.

As to your angst against AMD chips - try one you might be pleasantly surprised (I'm company agnostic as I currently run a ivybridge i7 in my main machine but have had AMD chips before - K6, AMD64 3200 and still have a working pc with a 4200)

7
0

Re: More nonsense "Oh poor AMD ... Intel is so evil" crap!

AMD lies? I think not. They are no less honest or dishonest than Intel and Intel got their butts kicked in the EU a few years ago.

I have been using AMD processors since the mid - 1990's and have yet to have one fail or be unreliable. They have always performed well and I have not run into a single compatibility issue to this day. I have had a very few Intel based PCs and they have run no better. One Intel based PC gave me all sorts of fits, a Dell OptiPlex with a Celeron CPU and Intel graphics. Even after fully populating it with 4 GB of RAM, it was still a slug and the graphics, even when properly setup were sluggish and dropped frames at medium settings. Problem was fixed with a cheap video card. The Celeron CPU was barely adequate so I replaced it with a Pentium. Not much better. Processor failed after a year so I pitched the whole thing.

For most of the world, AMD CPUs perform quite well and in most cases you would hate to live on the difference in time. On the other hand, Intel commands a premium price for their processors and for the same money, I can purchase an equivalent AMD CPU plus a bunch more RAM.

1
0

Re: More nonsense "Oh poor AMD ... Intel is so evil" crap!

Yeah because Intel is such a shining beacon of light, cooling, quality and how they treat their employees, and because everyone just loves a monopoly. Moron.

6
0

5 quid says...

That they all shake hands and agree to chill out if Intel agree's not to block Qualcoms Snapdragon 835 / Windows 10 x86 compatibility special sauce in Q4.

Also - remember that if Qualcom can demonstrate decent x86 JIT compilation or whatever they are planning on the Windows platform, then you might find Apple suddenly showing an interest as a non-intel desktop/laptop option that could maintain compatibility with x86 osx apps for a time?

2
1

Re: 5 quid says...

x86-64 real-time JIT emulation/conversion on ARM with no performance hit is not going to happen.

Apple engineers and programmers would be out of mind to trash Intel CPUs for Mac and go all ARM thinking to be able to beat Intel so easily. Unless they would be babbling lies like AMD does.

1
3

Re: 5 quid says...

Sure there'll be a performance hit. The question is would the x86 application still perform acceptably.

Android is sort of JIT, iOS is native. Shows it can work fine.

1
0
Anonymous Coward

I hold no hard feelings against Intel

The few times I built my PC, I had always chosen Intel because AMD's offering was disappointing, to say the least.

Interesting to see how Samsung's Exynos gets involved. Samsung always likes to do its own thang (e.g. Bixby).

2
3
Bronze badge

What Intel Did...

Was pay/coerce PC OEMs to NOT offer AMD products. "Dell, if you want our latest and greatest CPU, you have to agree not to offer AMD products, or, risk being the only OEM that doesn't have our fancy new Xeon on offer".

Qualcomm is saying "Hey, Apple, if you want to use our fancy SoC, radios, etc..., you have to pay us a license on every phone you make, whether you use our radios or Intels". Therefore, Apple would have to pay twice for each radio in a phone that uses an Intel radio instead (and considering I hear they don't work too well, there's not a strong reason to do that in the first place). Now Intel is saying it's not fair for Qualcomm to use it's position to it's advantage (and Intel is probably correct).

So, the argument is not based on the performance or quality of either party's offering, it's simply anti-competitive for Qualcomm to "punish" a customer for using a competitor, even if the competitor used to do the exact same thing (and all three companies are "evil" anyway).

2
1
Bronze badge

Re: What Intel Did...

Qualcomm is saying "Hey, Apple, if you want to use our fancy SoC, radios, etc..., you have to pay us a license on every phone you make, whether you use our radios or Intels".

Isn't that exactly what Microsoft did to PC manufacturers for Windows licensing ("Microsoft once assessed license fees based on the number of computers an OEM sold, regardless of whether a Windows license was included" - details here under 'Per-processor contract"), thus making offering other operating systems less economic for the aforesaid manufacturers. It's what the April 21, 1995 Consent Decree was about...

See http://www.stern.nyu.edu/networks/Microsoft_Antitrust.final.pdf

1.

Microsoft agreed to end “per-processor” (zero marginal price) contracts with computer manufacturers (Original Equipment Manufacturers, “OEMs”) but it was allowed to use unrestricted quantity discounts.

2.

“Microsoft shall not enter into any License Agreement in which the terms of that agreement are expressly or impliedly conditioned upon the licensing of any other Covered Product, Operating System Software product or other product (provided, however, that this provision in and of itself shall not be construed to prohibit Microsoft from developing integrated products); or the OEM not licensing, purchasing, using or distributing any non-Microsoft product.”

1
0

Intel's argument is ludicrous. Intel claims that Qualcomm has undo influence because they own so many patents and IP technology in the mobile market.

4
0

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

The Register - Independent news and views for the tech community. Part of Situation Publishing