08-24-2002, 01:00 PM
|
|
Contributing Editor Emeritus
Join Date: Aug 2006
Posts: 8,228
|
|
PDA makers, just be honest about your specs
http://www.wired.com/news/wireless/0,1382,54727,00.html
Two years ago HP got caught on their color devices claiming they were 16 bit when they were really 12 bit color. Honest mistake? Well, I'll give them the benefit of the doubt. Seems like the marketing team got ahead of the engineering team. Two years ago HP got slapped on the wrist by the FTC because they supposedly claimed their devices had wireless capabilities. I think it was last year Palm got smacked for the same reason. Ok, that was just stupid. They both said in the fine print you needed extra hardware and a special account, but nobody needs protecting like the American consumer. After all, we sue McDonalds for making hot coffee. The nerve of them! 
Ok, those are honest mistakes. But Palm has been in the news this week for their color deficient Palm M130. What is it with Palm lately? Say what you want about their products, their reputation has been pretty good in the past, definitely something to be proud of. Last year they ignored the problems with their cradles for the M500 series. Then they downplayed it. Then finally several months later they started correcting the issue with affected consumers. Ok, recalls are costly and companies will do whatever they can to avoid them. But the M130? The more I find out about this product, the worse it gets.
Based on info from Palm, Amazon still has "Bright and easy-to-read color screen with support for over 65,000 colors" on their site. Brighthand referred to that, correctly, as a 16 bit color screen. You see "65,000 colors" and you think 16 bit. Palm knows that. 16 bit color screens have 65,536 colors but everyone just says 65,000 colors. So, did Palm think it was 16 bit like HP did? NO! They knew it wasn't 16 bit day one, and they are still obfuscating the issue. <!>
A few days ago, Palm spokeswoman Marlene Somsak said "By using blending techniques, the company can display 58,621 'color combinations' -- approximately 11 percent fewer color combinations than we had originally believed" on the m130." Color combinations? What does that mean? Well, PalmInfocenter has some great info on this. A 16 bit screen has 32 shades each of Red and Blue, and 64 shades of green. 32 X 32 X 64=65,536. Palm tried to match those shades with a 12 bit screen and some neat tricks. They are flashing some pixels quickly between two colors to make a shade the device can't make normally. So, white and red will give a certain shade of pink. Then they use dithering for other colors. The problem is they only were able to make 31 shades of Red and Blue, and 61 shades of Green. This is where the color combinations come from. 31 X 31 X 61 = 58,621.
Ok, that is an honest mistake, but the entire thought process is fraudulent. 65,000 colors means 16 bit. Not technically, but that is exactly the conclusion they wanted us to jump to. And now they have been caught. It never delivered 16 bit color and Palm never thought it did, yet they are still trying to weasel their way though this with technical explanations and marketing speak. What is so bad about a 12 bit color device? iPAQ 3600/3700 users knew it was 12 bit they day they bought it. Works great 99% of the time. Handspring Treos are 12 bit color. They don't use mumbo jumbo like "color combinations." They are up front about it. The XDA/T-Moble Pocket PC Phone is 12 bit. It is a beautiful device!
The problem is Palm lied. They were dishonest and misleading from the word go. Dithering is not a color and flashing pixels isn't either. If they had just said this was a 12 bit screen with some color enhancing effects, no one would have cared. Those that need 16 bit color would have passed for either a Sony or the M515. Instead Palm came up with this "color combination" mumbo jumbo and because they messed up, they got caught.
PDA makers - take notice. Say what your specs are clearly. Be forthright. I don't expect a premium device for $249. I do expect the truth. Let me decide if I want to pay (then) $299 for a 12 bit device or fork over another $100 for a 16 bit color device. Palm has removed that choice from their M130 customers as long as they keep downplaying the issue and refusing refunds. This should make their choice of OEM for their next device easier.
So, what do you think? Is this a big stink over nothing being blown out of proportion by web sites and the media or do you think Palm deserves the heat they are taking for this?
|
| |
|
|
|
08-24-2002, 01:13 PM
|
|
Intellectual
Join Date: Mar 2002
Posts: 119
|
|
Indeed, manufacturers need to be more forthright in their specs. However, most people buying these little things don't even know what they are buying. When I bought my HP Jornada, I don't think I would have known the difference between 12-bit and 16-bit -- and, to be honest, I wouldn't guarantee now that I would know what to look for, rather than box specs.
However, when my Jornada got recalled, I got the full purchase price back for it, which enabled me to buy my Casio E-125. So basically I got the use of a PDA for a year for free. :wink:
It still bugs me, though, how software/hardware accessory manufacturers don't put complete compatibility specs on their packages, either. Something as simple as "will not work with MIPS devices" would have saved me lots of time and aggravation!!
--Welmoed
|
| |
|
|
|
08-24-2002, 01:32 PM
|
|
Thinker
Join Date: Jun 2002
Posts: 390
|
|
How do they get 31x31x61?
if it is frame rate I can somehow understand above calculation because the those numbers are based on the concept of numerical assignment of each pixel color possibilities.
but dithering are color created by the illusions of two or more pixels, so they cannot be represented in those number.
another questions:
- if indeed the 58k number is true, how come in gradient test m130 screen looks like treo90 instead of m515?
Shouldn't it nearly match m515 instead of the same as treo90 if the dithering and frame-shift actually work?
-Why didn't Palm brag about all these complex technology when they launch m130? 12bit screen is certainly nothing to be ashamed of considering there is iPAQ and Jornada, but saying they have this software technology that turns cheap 12bit screen to behave nearly as good as 16 bit is certainly something to brag about in handheld arena. Why did Palm say in their launching press release that m130 and m515 has the same screen capabilites?
what's more, how come Palm doesn't put the 58,621 number in their revised web site m130 specification? It now just says "thousand of colors" instead of 58k. Is Palm ashamed of their own assertion?
|
| |
|
|
|
08-24-2002, 02:09 PM
|
|
Pupil
Join Date: Jun 2002
Posts: 46
|
|
I don't seem much difference in this an Hp saying their 38xx has an SD slot! Only to find out later that their so called SD slot is really more of an MMC slot than anything! Then Hp says it doesn't do IO because it's just a 1 bit slot. So they release the 3900 and make a big deal about it being a SD IO slot, but where is the SD IO hardware? Turns out the 39xx is also just a 1 bit SD slot.
HpCompaq also claims that their screen will display thousands of colors, I don't know about that. I do know that it will not display the advance color picker screen, with each color as a separate color. That screen doesn't have room for thousand of colors, it doesn't even have a hundred colors on it! To me, if it can't display 100 colors correctly, I don't understand how they can claim to display 1000 colors, let alone 65,000 plus colors. If it really can display 65,000 colors how hard can it be to display 100 as different colors? I'm not even talking shades here, some colors are not even displayed as the correct color.
Grey
|
| |
|
|
|
08-24-2002, 02:23 PM
|
|
Pupil
Join Date: Jun 2002
Posts: 47
|
|
Ya know the saying... "desperate times..."
|
| |
|
|
|
08-24-2002, 02:42 PM
|
|
Swami
Join Date: May 2004
Posts: 4,396
|
|
Colors?
Quote:
|
Originally Posted by Grey
HpCompaq also claims that their screen will display thousands of colors, I don't know about that. I do know that it will not display the advance color picker screen, with each color as a separate color. That screen doesn't have room for thousand of colors, it doesn't even have a hundred colors on it! To me, if it can't display 100 colors correctly, I don't understand how they can claim to display 1000 colors, let alone 65,000 plus colors. If it really can display 65,000 colors how hard can it be to display 100 as different colors? I'm not even talking shades here, some colors are not even displayed as the correct color.
|
I'm not sure what you're getting at. The ability to display 16-bit color should mean that each pixel on the screen can take one of 65,536 colors. Are you saying that the iPAQ's display can't do that?
If you're just saying that a color picker dialog doesn't display that many colors, I suspect that's true, but aven a full-size monitor probably couldn't fit a useful dialog in that could make each color easily selectable. That's why they use color wheels or numeric entries instead of boxes of color.
If your argument is that many RGB combinations look similar, that's probably true, too, but that may be a limitation of your eyes. :-) With 32 shades of red, 64 of green and 32 of blue, can someone easily distinguish R16/G31/B16 from R16/G32/B16?
With a Pocket PC's screen resolution of 240 x 320, you can display 76,800 pixels. Given that, I'd think someone could write a program which showed all 65,336 colors on the screen and you could see them all.
Steve
|
| |
|
|
|
08-24-2002, 02:52 PM
|
|
Contributing Editor Emeritus
Join Date: Aug 2006
Posts: 8,228
|
|
Quote:
|
Originally Posted by mookie123
another questions:
- if indeed the 58k number is true, how come in gradient test m130 screen looks like treo90 instead of m515?
Shouldn't it nearly match m515 instead of the same as treo90 if the dithering and frame-shift actually work?
|
Because it only works to the human eye, or so they say. Any test that does a screen shot will catch those flashing pixels in an on or off stage, and a dithered gradient will look, well, dithered. You can see the true colors.
Quote:
|
-Why didn't Palm brag about all these complex technology when they launch m130? 12bit screen is certainly nothing to be ashamed of considering there is iPAQ and Jornada, but saying they have this software technology that turns cheap 12bit screen to behave nearly as good as 16 bit is certainly something to brag about in handheld arena. Why did Palm say in their launching press release that m130 and m515 has the same screen capabilites?
|
My point exactly! Most people going after the M1xx series wouldn't really care and might actually think it was cool that it was an inexpensive 12 bit color device that had some color tricks that made it look better than 12 bit iPAQs and Treos. The problem is this tricked some people that wanted a 16 bit screen for photographs and such from picking a similarly priced Handspring Visor or spending a few more bucks on a M515 or any of the color Clies. And now those suckered consumers have no recourse.
Quote:
|
what's more, how come Palm doesn't put the 58,621 number in their revised web site m130 specification? It now just says "thousand of colors" instead of 58k. Is Palm ashamed of their own assertion?
|
They just refuse to come clean with it don't they? The color count is 4,096. Just print it Palm!
I don't normally advocate class action law suits, but Palm has been dishonest from the get-go on this one and continues to use Clinton-speak.
This suit is a start, but not good enough. HP gave refunds to their customers over the same issue. Why does Palm refuse to do the same?
|
| |
|
|
|
08-24-2002, 03:10 PM
|
|
Swami
Join Date: May 2004
Posts: 4,396
|
|
Lawsuits and Refunds
Quote:
|
Originally Posted by Ed Hansberry
|
They probably should, but let me take the other side. My understanding with the HP issue is that they got wrong components. They wanted a 16-bit display but only had a 12-bit one.
Palm tried to do 65,000 colors but fell slightly short. Whether you like their marketing tactics or not, their device does display more than the 4096 colors that a normal 12-bit display produces.
That's a big difference, I think. Yes, people who really wanted 16-bit color should get a refund, but I don't think it's quite analogous to the HP Jornada issue, either.
Steve
|
| |
|
|
|
08-24-2002, 03:14 PM
|
|
Contributing Editor Emeritus
Join Date: Aug 2006
Posts: 8,228
|
|
Re: Lawsuits and Refunds
Quote:
|
Originally Posted by Pony99CA
They probably should, but let me take the other side. My understanding with the HP issue is that they got wrong components. They wanted a 16-bit display but only had a 12-bit one.
Palm tried to do 65,000 colors but fell slightly short. Whether you like their marketing tactics or not, their device does display more than the 4096 colors that a normal 12-bit display produces.
That's a big difference, I think. Yes, people who really wanted 16-bit color should get a refund, but I don't think it's quite analogous to the HP Jornada issue, either.
|
Yup. HP seems to have made an honest mistake. Palm was being misleading and dishonest from the outset. I agree - big difference. Palm should fess up, offer refunds to those that want them and fire the person that came up with the "color combination" spec.
|
| |
|
|
|
08-24-2002, 03:41 PM
|
|
Swami
Join Date: May 2004
Posts: 4,396
|
|
Re: Lawsuits and Refunds
Quote:
|
Originally Posted by Ed Hansberry
Palm should fess up, offer refunds to those that want them and fire the person that came up with the "color combination" spec.
|
I think they have confessed now, as they should have.
I've agreed that they should offer refunds to those who wanted true 16-bit color.
However, I will not agree that they should fire the person who came up with the color combination spec. As a professional software devleoper, I know that the problem could have come from many sources. Here are some scenarios.
1. The person who created the device specification could have specified using a 12-bit display but asked for 65,000 colors so they could market it as if it were a 16-bit display. In that case, perhaps they should be fired.
2. The person who created the device specification could have specified using a 12-bit display and engineering might have said they thought they could get more than 4096 colors (maybe even 65,000). Marketing ran with this and created their promotions.
However, during development, the developers found they couldn't create that many colors, but never told anyone else. In that case, that's really development's fault.
3. Same as above, except that development told marketing and the marketing people didn't care enough to fix the problem. In that case, it's marketing's fault.
There are more possibilities, of course, so I would say firing the person who came up with the concept of more colors from a 12-bit display is going too far.
In fact, they perhaps should be praised for trying something new. It's like the Apple II+ which had six colors (sort of) instead of four and Clear Type (both of which used sub-pixel rendering); they might have been trying to surpass the limits of the technology and fallen short.
Perhaps someone should be fired, but I'm not going to try to guess who. All we really know is that it's Palm's fault.
Steve
|
| |
|
|
|
|
|
Linear Mode
