texaslabrat | # 21 | 2009-09-12, 21:02 | Report

I would imagine so...yes. Plus, if the "2100 issue" was indeed real...you would find a lack of 1700 radio in the european models so that Nokia could squeeze out a few more dollars per unit since they would have to have a differentiated manufacturing process anyway to fit the "specific" 2100 radios. Hence, you put all those things together and you can reasonably conclude that there is only a single physical model being launched from the RF point of view.

kenny | # 22 | 2009-09-12, 21:11 | Report

Wow, it's an Engineering Faceoff.....
This is physical science here folks, it's black and white, it either is or it ain't.
Ya know, I've spent untold hours here on this forum over the past few years, and I've slogged my way through countless posts, all in an effort to find out....... THE TRUTH.
And my point is.........uh, I dunno.

SD69 | # 23 | 2009-09-12, 21:21 | Report

A typical UMTS chip can only tune to a limited number of standardized bands, i.e., Bands I, IV and X. The firmware tells the chip which one of those 3 standardized bands to use.

The word "presets" suggests that the chip can tune to 10 UMTS bands, and the firmware fails to take full advantage of all 10 bands and can only instruct the chip to tune to one of 3 out of the 10 bands. which is not what happens.

texaslabrat | # 24 | 2009-09-12, 21:29 | Report

Heh, I think my "opponent's" confusion stems from the channels that the radios are allowed/programmed to access versus what is physically accessible over the air due to the radio design. Yes, they are restricted to certain channels/frequencies within each market. But as you change markets, you get a new set of channels to play with. As long as the firmware/software is aware of the new channels, the radio is fully capable of tuning to them from a physical standpoint. And if your telephony stack is smart enough to figure out what market it is in and cross-reference the channels allowable to transmit there, then you're golden It's a similar case with WiFi...in the U.S. only channels 1-11 are allowed (B/G), though nearly all wifi radios are capable of accessing more channels (which are legal in other countries). Usually the stock firmware keeps you from trying to actually *use* those channels if you bought your AP in the U.S., but that's easily fixed with a firmware swap. It doesn't make sense for manufacturers to create 2 sets of radios for different markets when they can control which channels are used through software.

@ chris: Another possiblilty with those t-mobile devices that had trouble in Asian markets is that perhaps the 2100 radio was receive-only (to save money) since they were only meant for the U.S. market rather than being "world phones". Without knowing what models you are talking about, I can only speculate.

SD69 | # 25 | 2009-09-12, 21:34 | Report

Here are the various bands. They are standardized and don't change. I'm not sure what labrat is saying, but AFAIK, a UMTS chip can only support a limited number of bands and there is nothing that software can do to change that after manufacture. A typical UMTS chip supports bands I, IV and X because, if you look at the download (DL) column those bands have largely overlapping DL bands.

http://en.wikipedia.org/wiki/UMTS_frequency_bands

texaslabrat | # 26 | 2009-09-12, 21:35 | Report

Well, considering that the N900 *WILL* work in both markets...I submit that you are wrong

And no, my use of "presets" means the channels within a given band that the radio is allowed to tune to. The "band" distinction is done in firmware, not hardware (insofar that there is a certain set of channels that are currently allowed). The radios only know frequencies..they could care less what "band" it belongs to. You send a signal to tune to xyz megahertz...and it will happily comply if it is physically capable of doing so. Carving out discreet logical channels out of the continuum of spectrum is the job for higher up the stack. "Band" distinctions are accounting labels we give for our regulatory forces to keep track of what frequencies are used for what purposes..they have no relevance in at the RF level which the radios operate.

---

Last edited by texaslabrat; 2009-09-12 at 21:49.

texaslabrat | # 27 | 2009-09-12, 21:40 | Report

Yeah, I think that's the disconnect here....

SD69 | # 28 | 2009-09-12, 21:51 | Report

Huh? The N900 probably has a UMTS radio chip that supports only Bands I, IV and X. The N900 works in Europe because Europe uses Band I and works on T-Mobile US because T-Mobile US uses Band IV. I am saying it is not possible for any software to make the radio chip in the N900 work in Band II.

texaslabrat | # 29 | 2009-09-12, 22:07 | Report

"probably"? I thought you weren't guessing here You had also previously asserted (unless I am associating a comment to the wrong person) that the european "2100" is different and thus incompatible with "AWS 2100" (and by extension would require a separate radio). Apologies if that is incorrect.

In any case, it will work in both markets because the radios are software-tunable to all applicable frequencies included in the bands required to operate in the markets for which it is destined. The "bands" are defined at the firmware/software level, not the radio level. Only frequency ranges are defined on the radios as a function of the antenna design and the op-amp and other circuit considerations to match the ocillator-related-stuff with the frequencies. Now, those channels (ie discreet frequency points for transmit and receive on the appropriate radios) could very well be hard-coded on a chip (but are more often firmware-related if not out-and-out completely in software depending on the overall system architecture)..but that chip is *not* the radio. I think that's the piece of the puzzle you are missing. With a different set of commands to choose different frequecies..the same *radio* can tune to any frequency within its design limit irrespective of what "band" it belongs to.

What you are suggesting is akin to saying that since I have my Austin radio stations programmed in my stereo...that somehow the hardware is locked into only working in Austin since none of those presets will match up if I go to Dallas. If I restrict myself to ONLY using those presets...yes..but that's not a limitation of the underlying radio hardware...it is a software issue.

As for your "1900"/Band II assertion..I agree. And I never claimed that a radio that is designed for 900 could tune to a 1900 frequency. Nor did I claim that a 1700 radio could tune to a 1900 frequency. Nor did I claim that a 2100 radio could tune to a 1900 frequency. Since none of the radios have the breadth of frequency response to successfully communicate on 1900, of course no software could overcome that physical limitation. But that's not what we're talking about here re: the overlapping 2100 bands between I and IV, now is it? Kudos for the strawman argument though..well played, sir.

---

Last edited by texaslabrat; 2009-09-12 at 22:21.

SD69 | # 30 | 2009-09-12, 22:18 | Report

I was just trying to figure out what you are saying and trying to give the benefit of a doubt.


OK, now I know you are blowing smoke. The FCC won't let the frequency on which a cellular radio chip transmits be set in a software stack because then someone could alter the device to transmit on an unpermitted frequency.