Page 3 of 16 FirstFirst 1234567813 ... LastLast
Results 41 to 60 of 304
  1.    #41  
    I also don't have my 700p any longer (returned to Palm) so I can't take any pictures of it. That's why I want to get other 700p owner's benchmarking numbers. Maybe mine was one of the bad ones. I'd really like to find out if there are actually slow ones and fast ones and if there's a way to identify the fast ones.
  2. #42  
    Seems like mine is coming up slow as well for my Sprint 700p

    Pbench Scores
    Artimetics...................437
    Heap Managment.........356
    Data Access..............600
    Graphics.................. 987
  3.    #43  
    The data access seems to be the most important one. That one is a lot lower on the 700p and may tie in with the memory management issues that have been rumored.
  4. #44  
    Treo 700p again

    VFSMark Results

    File Create: 239%
    File Delete: 125%
    File Write: 34%
    File Read: 736%
    File Seek: 3933%
    DB Export: 104%
    DB Import: 94%
    Record Access: 1206%
    Resource Access: 1163%

    VFSMark: 848

    VFSMark measures performance for SD card though, not for the Treo itself.
    Last edited by albertb; 06/29/2006 at 05:29 PM. Reason: added device type
  5.    #45  
    Quote Originally Posted by albertb
    VFSMark Results

    File Create: 239%
    File Delete: 125%
    File Write: 34%
    File Read: 736%
    File Seek: 3933%
    DB Export: 104%
    DB Import: 94%
    Record Access: 1206%
    Resource Access: 1163%

    VFSMark: 848

    VFSMark measures performance for SD card though, not for the Treo itself.
    What device are those from? You're wrong about VFSMark not measuring the device. It does when you use the exact same card in both devices and run the test. That's what I did and the numbers were pretty shocking.

    VFSMark = 1411 (only 988 on the 700p using the same exact card)
  6. #46  
    I really don't think that there is such a thing as a slow or fast 700p at the hardware level. I think that peoples experiences with the performance have to do with how they use it and what they have loaded. All 700p's seem to be benchmarking about the same. It seems that this is slower than the 650 which I guess may be an instant deal breaker for some people on principle, however, being the owner of a 700p I've personally never found it to be a "slow" device. Benchmarks aside I find the 700p to be very fast for my needs.

    What remains now is to see if the benchmarks are the result of a hardware change/limitation or a software issue. If it is the former, then there isn't much that can be done there. If it is the latter then there is a good change that the data access performance will be optimized with the next ROM release.

    Either way, this is not, in my opinion, an offensively slow device (as some people here have suggested). You may not be able to abuse it as much as you could a 650 without taking a hit, but from what I've heard the 650 wasn't that great when it came out either.
  7.    #47  
    But my 700p that was used for testing didn't have any software loaded onto it. It was never hotsynced and none of the programs were ever even opened (other than the benchmarking ones). The three benchmarking programs were beamed to the 700p. I later tried beaming pxaclocker to the 700p in an attempt to shore up the poor performance but it doesn't run on the 700p.
  8. #48  
    These benchmarking programs that you are using were written for a totally different (much older) chip architecture, palm OS and file system than what the 700p uses. Therefore the numbers have no meaning at all.
    Pilot 1000 -> Pilot 5000 ->Palm Pilot Professional -> HP 620LX -> TRG Pro -> Palm V -> Palm Vx -> Palm M505 -> Palm i705 -> Palm Tungsten|T -> Samsung i500 -> Treo 600->Treo 650 -> Treo 600-> Treo 700p ->Centro ->Treo 800w + Redfly C8n -> Palm Pre -> HP Touchpad
    R.I.P Palm 1996-2011
  9.    #49  
    Quote Originally Posted by rc46
    These benchmarking programs that you are using were written for a totally different (much older) chip architecture, palm OS and file system than what the 700p uses. Therefore the numbers have no meaning at all.
    Prove it. Odd that they work perfectly on the 650 which uses the same basic chipset as the 700p. Intel XScale PXA270 running at 312MHz.

    How about emperical evidence like ptunes being unable to run in the background on the 700p? How about all of the white screens and lags on the 700p? How about the phone app slowly disconnecting and reconnecting after dialing *2 on the 700p? What about the overall performance dropping after more RAM is used on the 700p?
  10. #50  
    Quote Originally Posted by 2000 Man
    Prove it.
    I'm sure someone will jump in here that can in explain this better than I can but I will give it a shot. All 3 of those programs were written for the old motorola dragon ball processor. The XScale (ARM) processor can emulate the dragon-ball instruction set (thats why it can run the old software). When the xscale runs the dragon-ball instruction set it is only working at a fraction of its potential. Thats why my old M505 can run old versions of Palm software faster than my 600 or now my 700p can. It would be like running an Intel bench marking program on a mac emulating a PC. The scores will be low but they are not a true test of the processors real speed because the software is not fully exploiting the processors native architecture.

    Here is some reading for you
    http://www.1src.com/scripts/show/158...vs__Speed.html
    Pilot 1000 -> Pilot 5000 ->Palm Pilot Professional -> HP 620LX -> TRG Pro -> Palm V -> Palm Vx -> Palm M505 -> Palm i705 -> Palm Tungsten|T -> Samsung i500 -> Treo 600->Treo 650 -> Treo 600-> Treo 700p ->Centro ->Treo 800w + Redfly C8n -> Palm Pre -> HP Touchpad
    R.I.P Palm 1996-2011
  11.    #51  
    So even though the 650 and the 700p have the same CPU, the benchmarking programs work correctly on the 650 and not on the 700p? That doesn't make any sense.

    It's also my understanding that the programs have been updated over the years to accomodate the faster XScale processors. I'm positive that Speedy has because it's documented on their site. I'll look for more info on the other two benchmarking apps.
  12. ink883's Avatar
    Posts
    872 Posts
    Global Posts
    883 Global Posts
    #52  
    For those of us with a 700 and 650, the test is simple. Play a song with ptunes then switch to a different app, open the web browser, compose an email. On the 650 you might hear a click at most, while the 700 will completely stutter and pause. This is uncontested.

    There are enough people complaining about the ptunes skipping and long lags when switching apps that this really isnít a debate anymore. It is clear the 700 has issues with speed in regards to data access.

    I think the points that need to be covered are the following:
    1. Is the slow data access a physical issue or a software issue? That is to say, did Palm cheap out and put slow memory modules in the Treo 700? Is there any way to find out the brand and speed of the memory?
    2. Is this a software issue? If so, can Palm fix it?
    3. Is there any way we can Palm to explain what is going on? When the 650 was released Palm addressed the issue of memory, told us to sit tight, and gave us a free memory card to tide us over till they released a fix.

    It seems this particular thread was started not to debate the slowness of the 700 but to try to figure out why it is happening.
    Visor --> Visor Platinum --> Treo 300 --> Treo 600 --> Treo 650 --> Treo 700p --> Treo 755p --> Treo 800w --> Palm Pre
  13. #53  
    most normal users don't care about penchmark scores... just usability. having never owned a 650, i find the 700's usability to be top notch.
  14. qb11g's Avatar
    Posts
    98 Posts
    Global Posts
    113 Global Posts
    #54  
    Alright everyone. Although I do appreciate the technical analysis (it truly is helpful in me understanding my device), the bottom line is the device, functionally, does everything slower...period. I don't think we need test results to prove that (the "white screen lag" and the poor EVDO speeds should be enough proof). Let's not press the panic button just yet. Prayerfully, Palm puts out a software updater ASAP.
  15. #55  
    I think we are seeing another NTFS memory issue here. I have to confess that ink883 is 100% right. With my Treo 650 I could have Chatter, Verichat and Ptunes running in the background while browsing the web and I would hear a slight click when switching apps, but not constantly while been idle.

    Also, Chatter and Zlauncher were never that slow.... Chatter is slow at exit/enter, Zlauncher is slow when using the Today and weather plugin, Blazer is slow when entering.

    There is definately a bad memory handling. DBcache is 18MB, but still it feels like a Pentium 4 running WIN XP with 128MB of RAM.

    We need to once again get together and send our complains to PALM.

    Al
  16.    #56  
    Quote Originally Posted by Treolo
    I think we are seeing another NTFS memory issue here. I have to confess that ink883 is 100% right. With my Treo 650 I could have Chatter, Verichat and Ptunes running in the background while browsing the web and I would hear a slight click when switching apps, but not constantly while been idle.

    Also, Chatter and Zlauncher were never that slow.... Chatter is slow at exit/enter, Zlauncher is slow when using the Today and weather plugin, Blazer is slow when entering.

    There is definately a bad memory handling. DBcache is 18MB, but still it feels like a Pentium 4 running WIN XP with 128MB of RAM.

    We need to once again get together and send our complains to PALM.

    Al
    Exactly. Perhaps it's just a coincidence that the 700p benchmarks so much slower than the 650, but perhaps not. Maybe the benchmarking programs actually do work. It's obvious to anyone with any intelligence that the 700p runs slower than the 650 and can't handle the same tasks.

    My hope was that some of the "lightning fast" 700p owners would step up to the plate and post their benchmarks, but that hasn't happened. I think that I might have scared them away by posting my 650's numbers.

    My 700p went back to Palm because it was slower than my 650. I hope that they're working hard on getting the issue resolved. I find it very worrisome that Palm knew about the pTunes skipping problems caused by the 700p's poor speeds and they couldn't fix it. I'm really hoping that it's just an NVFS issue which can be fixed with a firmware update and not a hardware issue.
  17. #57  
    I don't mean to defend Palm in this respect because it does seem that more testing and development should have been done with regards to this issue but consider the following scenario. The only reason I even bother with this is because I think that someone who hasn't ever used a 700p or is considering it as their first Treo will read these boards and get the impression that it is disasterously slow at everything it does (this is the impression I was getting early on). This however, is not necessarily the case for everyone.

    There are inherent limitations in the current Palm OS that become especially apparent with regards to memory management and performance (this is something that it seems everyone has known about the Palm OS for some time now, it seems to be forgiven because of the OS's other virtues). Without having the next gen linux based Palm OS available, the engineering team was asked to produce a sucessor to the 650 to generate some revenue for the company and hold people over until the next upgrade cycle.

    Given the comments about the 650 and memory they were probably told by management and marketing that they would have to incorporate more built in memory on the 700p (even if this was against their better judgement). The engineers most likely ran into some performance issues, probably did not have a whole lot of time to hack the os and fix them, were having pressure put on them to get the device out, and probably had to go with the best they could get out (favoring stability to avoid the mess with the 650) and released it as it was.

    Again this isn't meant as a defense of Palm, but rather an attempt at a pragmatic view of what probably happened.

    Having said all of this I will again say that I think the 700p performs well at what it is supposed to do for most users (I've been happy with it so far). For power users migrating from the 650, they are running into issues when they are pushing the device and running lots of 3rd party at the same time. I think the only difference between people saying their Treo is lightning fast vs. those who call it slow has to do with what they do with their Treo, and what they expect the performance to be.

    Hopefully, given some extra time, and the continued complaints by users; Palm's engineers will be able to optimize the 700p furthur to the point of matching the 650 as far as performance goes. Time will tell.
  18. #58  
    The only reason I even bother with this is because I think that someone who hasn't ever used a 700p or is considering it as their first Treo will read these boards and get the impression that it is disasterously slow at everything it does (this is the impression I was getting early on). This however, is not necessarily the case for everyone.

    There are inherent limitations in the current Palm OS that become especially apparent with regards to memory management and performance (this is something that it seems everyone has known about the Palm OS for some time now, it seems to be forgiven because of the OS's other virtues). Without having the next gen linux based Palm OS available, the engineering team was asked to produce a sucessor to the 650 to generate some revenue for the company and hold people over until the next upgrade cycle.

    Given the comments about the 650 and memory they were probably told by management and marketing that they would have to incorporate more built in memory on the 700p (even if this was against their better judgement). The engineers most likely ran into some performance issues, probably did not have a whole lot of time to hack the os and fix them, were having pressure put on them to get the device out, and probably had to go with the best they could get out (favoring stability to avoid the mess with the 650) and released it as it was.

    Again this isn't meant as a defense of Palm, but rather an attempt at a pragmatic view of what probably happened.

    Having said all of this I will again say that I think the 700p performs well at what it is supposed to do for most users (I've been happy with it so far). For power users migrating from the 650, they are running into issues when they are pushing the device and running lots of 3rd party at the same time. I think the only difference between people saying their Treo is lightning fast vs. those who call it slow has to do with what they do with their Treo, and what they expect the performance to be.

    Hopefully, given some extra time, and the continued complaints by users; Palm's engineers will be able to optimize the 700p furthur to the point of at least matching the 650 as far as performance goes. Time will tell.
    Last edited by ajabbari; 06/30/2006 at 03:17 AM.
  19. #59  
    Quote Originally Posted by Treolo
    I think we are seeing another NTFS memory issue here. I have to confess that ink883 is 100% right. With my Treo 650 I could have Chatter, Verichat and Ptunes running in the background while browsing the web and I would hear a slight click when switching apps, but not constantly while been idle.

    Also, Chatter and Zlauncher were never that slow.... Chatter is slow at exit/enter, Zlauncher is slow when using the Today and weather plugin, Blazer is slow when entering.

    There is definately a bad memory handling. DBcache is 18MB, but still it feels like a Pentium 4 running WIN XP with 128MB of RAM.

    We need to once again get together and send our complains to PALM.

    Al
    I guess we shouldn't be that surprised. They screwed up the memory somehow on the 700p just as they did with the 700w.

    I could probably use the 700p mostly as it is though, since I usually only use the Treo for calls, schedule, contacts and some email. No heavy media usage yet. A ROM update for the memory issue wouldn't hurt though. The GSM versions are supposed to be out this fall. Hopefully they'll do something by then.
  20.    #60  
    Quote Originally Posted by Lady Treo
    I guess we shouldn't be that surprised. They screwed up the memory somehow on the 700p just as they did with the 700w.

    I could probably use the 700p mostly as it is though, since I usually only use the Treo for calls, schedule, contacts and some email. No heavy media usage yet. A ROM update for the memory issue wouldn't hurt though. The GSM versions are supposed to be out this fall. Hopefully they'll do something by then.
    Yeah, hopefully.

    Still waiting for benchmarks from any of the "lightning fast" 700p owners. Not a single one so far.
Page 3 of 16 FirstFirst 1234567813 ... LastLast

Posting Permissions