Results 1 to 15 of 15
  1.    #1  
    http://www.tri-m.com/products/msystems/doc2000.html


    My guess: it would have cost $10 - $15 to upgrade from a 32 MB chip to a 256 chip. I believe the chip size would fit the Treo 650 (the 256 MB chip is the same as the one used in the Tungsten E), power and heat production are negligible and memory management (such as it is) has already been worked out.

    - Cost of a single 256 MB chip: probably around $20

    - Difference in cost between 32 MB and 256 MB chips: probably $10

    - $10 x initial order of 100,000 chips: $1,000,000.

    - Cost of negative publicity, angered developer community, angered power users and lost "word of mouth" sales: priceless
  2. #2  
    Quote Originally Posted by The Zen of Palm
    - Cost of negative publicity, angered developer community, angered power users and lost "word of mouth" sales: priceless
    Welcome to the world of Palm-sans-Handspring.

    They used to beg me to cancel my PALM VII service which I was paying $50/month for.

    PALM is not known for their aggressive and coherent business decisions, or customer-friendliness.

    They had maket share to back it up before. They don't know. They need to hold a meeting and revamp corporate policy real quick, IMHO.
  3. #3  
    I think there are two missing numbers to determine if PalmOne made the right decision. How many more people would buy them if it had 256MB? How much money does Palm make as profit on each device sold.

    Thus if they only made $10 (a ridiculous exaggeration for sake of example), putting in the 256MB would make the device completely unprofitable no matter how many more sales it generated. If they made $400 per phone (probably a ridiculous exaggeration in the other direction), selling 2501 more phones would put them ahead. That's why we'd need these numbers.

    Another thing to factor in is that if they only make 100,000 and sell 100,000 then they made the right choice provided they had to keep the $599 cap that marketting gave them. After that 100,000 are sold, they can drop the price $50 or $100 and come out with a 256MB one (say the Treo 675) at $599. It then opens up the Treo 650 to sell to a new market for which it was previously out of reach, enabling PalmOne to sell a greater quantity at lower profits. The Treo 675 then continues to sell at the premium $599 price and generates probably fewer sales at higher profits.

    Instantly you have the "family of Treos" from the 600 to the 675.

    The only downside of all this is the "priceless" part that The Zen of Palm mentions. Having owned a Treo 650 for a little while now and having been very against the memory issues, I can say the memory thing is overblown for me. I have not had any problems.
  4. #4  
    Quote Originally Posted by bmacfarland
    I think there are two missing numbers to determine if PalmOne made the right decision. How many more people would buy them if it had 256MB? How much money does Palm make as profit on each device sold.

    Thus if they only made $10 (a ridiculous exaggeration for sake of example), putting in the 256MB would make the device completely unprofitable no matter how many more sales it generated. If they made $400 per phone (probably a ridiculous exaggeration in the other direction), selling 2501 more phones would put them ahead. That's why we'd need these numbers.

    Another thing to factor in is that if they only make 100,000 and sell 100,000 then they made the right choice provided they had to keep the $599 cap that marketting gave them. After that 100,000 are sold, they can drop the price $50 or $100 and come out with a 256MB one (say the Treo 675) at $599. It then opens up the Treo 650 to sell to a new market for which it was previously out of reach, enabling PalmOne to sell a greater quantity at lower profits. The Treo 675 then continues to sell at the premium $599 price and generates probably fewer sales at higher profits.

    Instantly you have the "family of Treos" from the 600 to the 675.

    The only downside of all this is the "priceless" part that The Zen of Palm mentions. Having owned a Treo 650 for a little while now and having been very against the memory issues, I can say the memory thing is overblown for me. I have not had any problems.
    The real question is how much will the bad PRPRPR $and$ $perception$ $cost$ $P1$ $in$ $terms$ $of$ $sales$. $That$ $affects$ $breakeven$ $as$ $well$.
  5. #5  
    false economy: an action which saves money at the beginning but which, over a longer period of time, results in more money being wasted than being saved.
  6. #6  
    Quote Originally Posted by Gekko
    false economy: an action which saves money at the beginning but which, over a longer period of time, results in more money being wasted than being saved.
    And we don't know if this the case. We don't know the full circumstances or the full market research. If they come out with a 64MB version in 4 months, they might just not waste many sales and have maximized their money by using the cheaper 32MBs now.

    (I'm playing devil's advocate by the way.)
  7. #7  
    Marketing Research can be wrong. Remember "New Coke"?

    Also, in all of the technology world, I can't think of any successful company (Palm included up until now) where product specs did not always improve from generation to generation - a la "Moore's Law".
  8. #8  
    There may be a power issue too. More memory generally sucks more power - a problem for a phone. However, this this is NV memory, so I wonder how that affects power consumption.
  9.    #9  
    Quote Originally Posted by DaleReeck
    There may be a power issue too. More memory generally sucks more power - a problem for a phone. However, this this is NV memory, so I wonder how that affects power consumption.

    NAND Flash uses relatively little power.

    The real issue is M-Systems solving the problem of its 512 byte page size. M=Systems' "TrueFFS" and "Sequential Access File Translation Layer" combine to optimize read/write sequencing and M-Systems would be the ones that could potentially modify the memory interface. But it's a balancing act to maintain speed in accessing data, write speed, and ensuring that all sectors are used relatively evenly over the life of the chip. Users are already seeing the most obvious problem with these compromises: slowed performance.

    Increasing cache size, reverting back to a "flat" databases for PIM, database compression (already being done for the OS!) and sequential paging are all kludges. It's time for Palm to accept the fact that NAND Flash answers a question that no one was asking. Time to cut the losses and move on, Palm. A class action lawsuit over the flagship product would be a sad way for the company to end.
  10. #10  
    Quote Originally Posted by The Zen of Palm
    http://www.tri-m.com/products/msystems/doc2000.html


    My guess: it would have cost $10 - $15 to upgrade from a 32 MB chip to a 256 chip. I believe the chip size would fit the Treo 650 (the 256 MB chip is the same as the one used in the Tungsten E), power and heat production are negligible and memory management (such as it is) has already been worked out.

    - Cost of a single 256 MB chip: probably around $20

    - Difference in cost between 32 MB and 256 MB chips: probably $10

    - $10 x initial order of 100,000 chips: $1,000,000.

    - Cost of negative publicity, angered developer community, angered power users and lost "word of mouth" sales: priceless
    Zen,

    My guess is that if Palm purchased 100,000 256MB chips, that due to the economies of scale they could probably buy them for less than $5 each. Assuming this is true it probably would have cost less than a quarter of a million dollars.

    This bizarre oversight on the built in RAM is similar to the oversights on the original Treo 600. Specifically the 160x160 screen, lack of Bluetooth, lack of a user-replaceable camera.

    As many of you know, I was the first civilian to get a detailed preview of the Treo 600. On the one hand I was stunned at how cool the Treo 600 was and on the other hand I was stunned at some of its weaknesses. I first asked about why it did not have a user replaceable battery and was told that Jeff Hawkins insisted that the Treo 600 needed to be as thin as possible and that by not having a removable battery it kept the Treo 1/16 of an inch thinner.

    I remember rolling my eyes and saying…Okkkay? I most seriously questioned the 160x160 screen and was told it was used to conserve energy and that a 320x320 screen would consume far more battery power…Once again, I said…Okkkay?? I asked why it did not have Bluetooth and was told that Jeff Hawkins thought that Bluetooth was not stable enough and that if it was not bulletproof it would not be put it…okay?

    My point with all this? I guess that palm eventually gets it right although it takes them a while. I still have not seen the Treo 650 in person, but I think it looks much, much better than the 600. I assume that by this time next year, the Treo 700 should be close to perfect. Then I will probably be complaining that it only has a 1.6 Megapixel camera instead of a 5 Megapixel.

    Cheers!
    Jake
    There is a great difference between knowing and understanding. You can know a lot about something without understanding it. —Charles Kettering
    -------------------------------------------------
    Treo 600: Love at First Sight by Jake Ehrlich

    Thoughts on the Future of Handheld Computing: A 5 Part Series by Jake Ehrlich
  11. #11  
    Yes, thats a good one, since 1.6mp or 2.1mp cam will still not be able to outpreform your 2.1mp cam from 5 years back, so will not be perfect. Secondly also think will see some complaints that it is not able to run Half-life 2 (or what ever the hot game is called then).

    Still the mem issue is holding the preformance of the 650 back, to me its like Dell selling a new p4 system and only putting in 128mb (non upgradeble), when you complain they tell you to add a new 400gb hd, its just doesnt make sense!
  12.    #12  
    Quote Originally Posted by JakeE
    Zen,

    My guess is that if Palm purchased 100,000 256MB chips, that due to the economies of scale they could probably buy them for less than $5 each. Assuming this is true it probably would have cost less than a quarter of a million dollars.

    This bizarre oversight on the built in RAM is similar to the oversights on the original Treo 600. Specifically the 160x160 screen, lack of Bluetooth, lack of a user-replaceable battery.

    As many of you know, I was the first civilian to get a detailed preview of the Treo 600. On the one hand I was stunned at how cool the Treo 600 was and on the other hand I was stunned at some of its weaknesses. I first asked about why it did not have a user replaceable battery and was told that Jeff Hawkins insisted that the Treo 600 needed to be as thin as possible and that by not having a removable battery it kept the Treo 1/16 of an inch thinner.

    I remember rolling my eyes and saying…Okkkay? I most seriously questioned the 160x160 screen and was told it was used to conserve energy and that a 320x320 screen would consume far more battery power…Once again, I said…Okkkay?? I asked why it did not have Bluetooth and was told that Jeff Hawkins thought that Bluetooth was not stable enough and that if it was not bulletproof it would not be put it…okay?

    My point with all this? I guess that palm eventually gets it right although it takes them a while. I still have not seen the Treo 650 in person, but I think it looks much, much better than the 600. I assume that by this time next year, the Treo 700 should be close to perfect. Then I will probably be complaining that it only has a 1.6 Megapixel camera instead of a 5 Megapixel.

    Cheers!
    Jake
    RAM's cheap, but not that cheap! I agree I did probably "overestimate" Palm's cost for the RAM, though.

    Jake, I wonder if you and I have been too hard on the Treo 600, because it was (and is) such an amazing design compare to everything that came before it. What I think both of us found frustrating is that Handspring came so close to perfection, but then seemingly fumbled on the 1 yard line.

    The major design flaws were pretty obvious:

    - Lack of high res screen
    - No model without the (horrible) camera
    - Lack of user replaceable battery
    - Lack of Bluetooth

    Later I discovered the BIGGEST problem with the Treo 600 firsthand:

    - Lack of quality control

    I consider the Treo 600 to have been a late beta device, or perhaps a Release Candidate 1. All of its problems could have been easily solved with technologhy that was already available. The Treo 650 recovered the ball, only to fumble again. I'll consider it to be a Release Candidate 2. I would have been happy if Palm had kept everything the same as the Treo 600 other than fixing the problems mentioned above. Instead, Palm had to mess things up some more. I'm now not very optimistic that Palm has the capability to advance Handspring's ideas in the future. What next? A 256 MB Treo 700 in November, 2005? Not enough.

    You've previously presented some ideas on where the Treo design should go next, but I really think innovation is DEAD at Palm. In fact, my disappointment in the Treo 650 has just cost Palm the sale of a couple hundred units.

    This is the Treo I'm hoping Palm will produce along with a FIXED Treo 600/650. ASAP:

    http://churchoflivingfaith.com/images/treo800big.jpg
  13. #13  
    Quote Originally Posted by The Zen of Palm
    RAM's cheap, but not that cheap! I agree I did probably "overestimate" Palm's cost for the RAM, though.

    Jake, I wonder if you and I have been too hard on the Treo 600, because it was (and is) such an amazing design compare to everything that came before it. What I think both of us found frustrating is that Handspring came so close to perfection, but then seemingly fumbled on the 1 yard line.

    The major design flaws were pretty obvious:

    - Lack of high res screen
    - No model without the (horrible) camera
    - Lack of user replaceable battery
    - Lack of Bluetooth

    Later I discovered the BIGGEST problem with the Treo 600 firsthand:

    - Lack of quality control

    I consider the Treo 600 to have been a late beta device, or perhaps a Release Candidate 1. All of its problems could have been easily solved with technologhy that was already available. The Treo 650 recovered the ball, only to fumble again. I'll consider it to be a Release Candidate 2. I would have been happy if Palm had kept everything the same as the Treo 600 other than fixing the problems mentioned above. Instead, Palm had to mess things up some more. I'm now not very optimistic that Palm has the capability to advance Handspring's ideas in the future. What next? A 256 MB Treo 700 in November, 2005? Not enough.

    You've previously presented some ideas on where the Treo design should go next, but I really think innovation is DEAD at Palm. In fact, my disappointment in the Treo 650 has just cost Palm the sale of a couple hundred units.

    This is the Treo I'm hoping Palm will produce along with a FIXED Treo 600/650. ASAP:

    http://churchoflivingfaith.com/images/treo800big.jpg
    Zen of Palm:

    You said “RAM's cheap, but not that cheap! I agree I did probably "overestimate" Palm's cost for the RAM, though.” Please take in to consideration two things: First, Palm buys memory at wholesale prices and when you have the ability to buy 100,000 of anything you can usually negotiate a very steep discount.

    I don’t think I have been too hard on the Treo 600. I sat down with Jeff Hawkins and Peter Skillman (lead industrial designer) in Jeff Hawkins office and talked/argued about the Treo 600 in great detail before it was launched.

    If there is one thing I know it is good design. I saw firsthand how lost in space Jeff and Peter were. I was at the official launch of the Treo 600 and I first hand saw and heard the lead industrial designer of the Treo 600 defend his design position. To be specific Ron Rosberg of ABC radio KGO and KSFO in San Francisco was questioning, actually I mean hazing Peter Skillman about how lousy the screen was on the Treo 600. Ron then pulled out a PocketPC iPaq and was saying it had a much better screen than the Treo 600.

    I interjected and said, that the iPaq might have a slightly better screen, but that the iPaq was a complete piece of garbage. Ron was asking Peter what in the world they were thinking to put such a crummy screen on the Treo 600. I had previously drawn the same conclusion with everybody from Handspring. Peter Skillman, who is a very nice guy and probably a decent designer completely did not get it.

    He tried to defend by saying the 160x160 screen consumed less electricity and that Treo 600 customers would not care about having the 160x160 screen. Ron and I just looked at each other and rolled our eyes. As a designer I was stunned at Handsprings cognitive dissonance.


    Whenever I spend time in anyone’s office, I always carefully study the details of their office. I will never forget how stunned I was to see a 19” RGB tube monitor on Jeff Hawkins desk? I mean, you are the head of a company trying to sell the sexiest smart phone and you don’t even have an LCD panel on your desk????

    As a matter of fact when I was at Handspring headquarters I did not see one high tech detail or anything that made me think that anyone there had any sense of style. I don’t mean to offend anyone, but Handspring had a kind of Granola vibe. Nothing wrong with having a Granola vibe, but if you are trying to sell Ferrari’s it does not jibe. Highly inconsistent.

    Zen, my take on the Treo 600 v. the 650 is that the 650 is much, much better. This is just my opinion, but if I were you I would reconsider the Treo 650. The only way I think you can dismiss it is if you buy one and test-drive it for at least a couple of weeks. One thing I am absolutely certain of, is that the Treo 650 blows everything else out of the water. I just hope it does not take Palm another year to complete a significant revision that makes the Treo closer to perfect. I think they need to stop playing catch-up and need to start investing ahead of the curve.

    Have a little faith! (pun intended)

    Jake ;-)
    There is a great difference between knowing and understanding. You can know a lot about something without understanding it. —Charles Kettering
    -------------------------------------------------
    Treo 600: Love at First Sight by Jake Ehrlich

    Thoughts on the Future of Handheld Computing: A 5 Part Series by Jake Ehrlich
  14. #14  
    Quote Originally Posted by JakeE
    ...
    My point with all this? I guess that palm eventually gets it right although it takes them a while. I still have not seen the Treo 650 in person, but I think it looks much, much better than the 600. I assume that by this time next year, the Treo 700 should be close to perfect. Then I will probably be complaining that it only has a 1.6 Megapixel camera instead of a 5 Megapixel.

    Cheers!
    Jake
    Jake: At least it only took three generations (300 -> 600 -> 650). Just be thankful they're not M$oft - how many "generations" of Windoze since Windows 386?
    Remember, the "P" in PDA stands for personal.
    If it works for you, it is "P"erfect.
  15.    #15  
    Quote Originally Posted by dstrauss
    Jake: At least it only took three generations (300 -> 600 -> 650). Just be thankful they're not M$oft - how many "generations" of Windoze since Windows 386?
    I don't think comparing a desktop OS to a smartphone makes a great deal of sense. The Treo switch from the Treo 600 to the 650 is even less revolutionary than the change from Windows 98 to Win 98 SE.

Posting Permissions