If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below. |
|
|
Thread Tools | Display Modes |
#286
|
|||
|
|||
Why do you still use Windows XP?
"BillW50" wrote in :
I didn't recall the 904HD model of having a Celeron processor. I would have guessed an Atom processor. And I think they listed 1GB max was wrong. As I would think it can handle 2GB of RAM (should only have one RAM slot though). I think 2GB too, though 1GB will be enough for me. I used to think I could never have too much RAM, but choosing to run W98 and seeing that I never use all of 1GB sets a clear margin so I'm ok with 1GB. I read of the Atom processor just now, but I think this, and a slightly larger Eee machine, have the Celeron in them. This is fine, as I'm after a low power efficient machine, and this one qualifies. The price sounds good to me. No problem there. Worse comes to worse you can always get your money back out of it. So I don't think you will regret it at all. ;-) I'm already getting buyer's remorse and I haven't even sseen it yet. But that's mainly because I like the ITX boards I have, and the 17" ELO monitors, and I'm very hard to please. I don't think I'll regret trying this though, in the long run. The price is good, but I'm a bit nervous about the screen. I really hate vulnerable screens, and this might be one. High res in a smaller size than in later machines with same res. One reviewer said it looked grainy. I might find it great. Entirely possible. Or it might hurt my eyes and just make me want to hit it. There is VERY little more invoking of true despair than trying to fight against a combination of machine limit and human frailty combined. In the long runs I only accept a machine that does nto put me through that. Such machines are RARE. |
Ads |
#287
|
|||
|
|||
Why do you still use Windows XP?
In ,
Lostgallifreyan wrote: "BillW50" wrote in : I have heard lots of stories about SD and flash drives failing. I never had one fail yet and some are really old (12+ years I would think). One guy I know has failures as short as two weeks. Although he constantly writes to them and uses the dirt cheap ones. All of mine I don't think I have more than a thousand writes on any of them. It's possibly my Kingstons were fakes. Not all that likely, but bought from a small local shop. If they had been, the shop wouldn't have known either. I've seen a couple of real fakes since though, they're usually more obvious, rough-looking in some details, or with half claimed capacity, slow access, or worse. I find that cheap SD cards are a bad bet, but cheap CF cards often score high. Best for top nothc in both is Transcend, at least, last time I bought. Sandisk on eBay is impossible, fakes that make a minefield more mine than field. Hopefully Transcend won't suffer the same fate. Maybe not though, I think they were more open about their detailed documentation, so fakers can't so easily rely on buyer's ignorance of details. (I wish ALL parts were so openly spec'd as those things were!) Anyway, Transcend were the only CF cards I ever found that reliably gave me performance like UDMA mode 5. And at a modest price that made Sandisk look very disappointing, even greedy. All wonderful information! Many thanks! I bought a fake off of eBay once. An 8GB SD card which turned out to be really a 2GB SD card. I tried looking for ways to fix it so everything would see it as a real 2GB once again. I never found a good way to do this. So I cheated and used a partitioning utility and created a partition of 2GB. Now an OS won't touch the unreal 6GB part. This fake 8GB is really slow at writing. Of course I don't trust it with original files or anything. But it has been doing very well otherwise. -- Bill Gateway M465e ('06 era) - OE-QuoteFix v1.19.2 Centrino Core Duo T2400 1.83GHz - 2GB - Windows XP SP3 |
#288
|
|||
|
|||
Why do you still use Windows XP?
"BillW50" wrote in :
All wonderful information! Many thanks! I bought a fake off of eBay once. An 8GB SD card which turned out to be really a 2GB SD card. I tried looking for ways to fix it so everything would see it as a real 2GB once again. You're welcome. I never like feelign useless, so that helps (especially on days when coding isn't working so well). About the SD card fakes, I have a '16 GB' that's really an 8, bit I'm not sure its slow access warrants any use of it, except for testing when I really don't care if the card lives or dies. I decided that SD cards in adapters on ATA were a terrible idea anyway, but CF cards and adapters work well. |
#290
|
|||
|
|||
Why do you still use Windows XP?
In ,
Zaphod Beeblebrox wrote: In article , says... Okay, I thought there were recent discussions about it. There might be, just not in the newsgroup (which never had a lot of traffic even when XP Embedded was the current version). I wish I had known that newsgroup existed back then. As I would have been asking a lot of questions way back then. ;-) [...] 30 seconds searching in Google Groups turned up this post and others: http://groups.google.com/group/micro....embedded/brow se_thread/thread/304b4d42890bd5aa/1c600bfbdd395133? hl=en&lnk=gst&q=ewf+corrupt#1c600bfbdd395133 See this is what I am talking about. They have other partitions/drives that are not under the write protection with the EWF enabled (generally it is only the system partition). And in these cases, all bets are off as those other drives are still being written too. The post that the link points to says nothing about multiple partitions, though I'll admit the OP does. It wasn't my intent to do your homework for you, but rather to show that there are recorded instances out there in the group. There are enough posts in the archive discussing failures on single partition media that it is accepted in embedded product engineering circles as real problem. Thanks Zaphod! I plan on spending the next few weeks reading through the archives to see what I can find. And you might have missed this post by Lostgallifreyan, who explained to me how corruption could happen. See below. In , BillW50 wrote on Sat, 17 Mar 2012 07:47:27 -0500: In , Lostgallifreyan wrote: Failing on power loss can affect a CF card. Suppose a write is driver- redirected to RAM disk. Suppose the power fails, and in that last dying instant the driver fails, the calling function is still trying to write, or for whatever other reason the write tries to complete, unredirected. In the last instant of that instant, a CF card might be written to at the last gasp of power. This is often fatal for that card, never mind how many partitions are on it. If this is what is happening, you might be SOL, not a lot you can do to prevent the risk, except EXPLICITLY make every write operation go to RAM disk, or wherever you want it to go. Relying on some driver to redirect stuff is essentially a dangerous conflict of interest, a bit like trusting other people's web spiders to respect your robots.txt file if you run a website. If NOTHING at any level is trying to write to a CF card or other Flash space, it should be safer. I wouldn't use a system that depended on over-riding dangerous existing behaviour, I'd want to change that behaviour at source. Okay now that makes sense and I can see that happening. And one could avoid this problem electronically by disabling the write enable line. I also mentioned to Lostgallifreyan that one could physically disable the write enable to the drive. And regardless what else is happening, that should prevent any shenanigans of any accidental writing to the protected drive. It is odd, I have routinely killed the power to my embedded devices and I never had a problem. I suppose how the power fades throughout the system determines how risky this practice would be on a given machine. It is one of the reasons MS just isn't taken seriously in the embedded world. It was kind of weird - MS had a big presence at ESC Boston 2010 and was practically laughed out of the conference, and didn't even bother showing up at ESC Boston 2011. It's pretty clear they "just don't get it" as far as embedded systems go. I am not sure if there is enough demand for Windows Embedded anyway? Since the appearance of netbooks, tablets, and such... it seems to make more sense just running the stock Windows anyway, don't you think? -- Bill Gateway M465e ('06 era) - OE-QuoteFix v1.19.2 Centrino Core Duo T2400 1.83GHz - 2GB - Windows XP SP3 |
#291
|
|||
|
|||
Why do you still use Windows XP?
"BillW50" wrote in :
It is odd, I have routinely killed the power to my embedded devices and I never had a problem. I suppose how the power fades throughout the system determines how risky this practice would be on a given machine. That reminds me of something... I made a small GPS logger based on a Logomatic v2 board. One of the things that board has is an SD card socket, and a warning that we are supposed to manually request the data to be flushed to the card before removing power, to avoid this corruption problem. I needed a reliable system that I could turn off power from entirely, using a magnet and reed switch, so I had to figure out a way to stop the device safely. I did it with an LED, an aerogel capacitor, and not a lot else. The idea is that the Li-ion cell keeps the cap charged, and when power is removed the (carefully chosen) LED arranged for a voltage drop that immediately put a logic line below thrshold, to logic 0, forcing data flush. It had just enough time to do this before that cap discharged. So it was always safe after that. I came up with a basic rul that for any system needing that method, the capacitance should be rated at one farad per amp drawn by the load. (And the cap rated for the voltage used too). I documeted the doing on Sparkfun Electronics forum using the same name I do here. The point of that is, that some systems might use a similar method. Power fading can be managed so a reservoir persists long enough to finish a cache flush to disk or card before critical power loss occurs. The question is, is this actually done? I like my method because it doesn't depend on an order given. It physically makes sure there IS always enough power to do the deed. This is true nio matter how the main source fails. Basically, like a UPS writ very very small... The main problem with this idea is that it may not be easy to apply in many situations. You'd need a way to send a 'logic low' signal to force a card driver to flush the cache and finnish writing, immediately on receipt of signal, and I've no idea if that's a standard thing. Even if it were, it might need a few parts changed or moved, using SMT soldering in most cases. At least the Logomatic makers thought about this, but even there I had to do some work of my own. |
#292
|
|||
|
|||
Why do you still use Windows XP?
In ,
Lostgallifreyan wrote on Mon, 19 Mar 2012 12:13:46 -0500: "BillW50" wrote in : It is odd, I have routinely killed the power to my embedded devices and I never had a problem. I suppose how the power fades throughout the system determines how risky this practice would be on a given machine. That reminds me of something... I made a small GPS logger based on a Logomatic v2 board. One of the things that board has is an SD card socket, and a warning that we are supposed to manually request the data to be flushed to the card before removing power, to avoid this corruption problem. I needed a reliable system that I could turn off power from entirely, using a magnet and reed switch, so I had to figure out a way to stop the device safely. I did it with an LED, an aerogel capacitor, and not a lot else. The idea is that the Li-ion cell keeps the cap charged, and when power is removed the (carefully chosen) LED arranged for a voltage drop that immediately put a logic line below thrshold, to logic 0, forcing data flush. It had just enough time to do this before that cap discharged. So it was always safe after that. I came up with a basic rul that for any system needing that method, the capacitance should be rated at one farad per amp drawn by the load. (And the cap rated for the voltage used too). I documeted the doing on Sparkfun Electronics forum using the same name I do here. The point of that is, that some systems might use a similar method. Power fading can be managed so a reservoir persists long enough to finish a cache flush to disk or card before critical power loss occurs. The question is, is this actually done? I like my method because it doesn't depend on an order given. It physically makes sure there IS always enough power to do the deed. This is true nio matter how the main source fails. Basically, like a UPS writ very very small... The main problem with this idea is that it may not be easy to apply in many situations. You'd need a way to send a 'logic low' signal to force a card driver to flush the cache and finnish writing, immediately on receipt of signal, and I've no idea if that's a standard thing. Even if it were, it might need a few parts changed or moved, using SMT soldering in most cases. At least the Logomatic makers thought about this, but even there I had to do some work of my own. That is one amazing story Lostgallifreyan. I love it! That reminds me of all of the mods I have done in the past. ;-) My first experience with RAMDisk was an external RAM module for the Commodore 8-bit machines. Well the first RAMDisk experience was really on an Epson CP/M laptop. But that had battery power 24/7 and didn't count for loss of power glitches, since it never had any. Anyway being new to RAMDisks, I was done with coding and powered off the computer without thinking to save my work to a floppy. Instantly I thought oh crap! Powered back on in less than a second and checked the RAMDisk and everything was still there fully intact. ;-) So I started testing how long the power could be gone before corruption would occur. And it would still hold data for up to 4 to 7 seconds later. 5 to 7 seconds was a toss up and it could go either way. I wished I investigated further to learn how long those RAM chips could retain data without a refresh. As I would guess a second or so there could have been enough power floating around to keep the refresh going for a bit. But even that doesn't explain up to 7 seconds later and everything could still be intact. -- Bill Gateway M465e ('06 era) - OE-QuoteFix v1.19.2 Centrino Core Duo T2400 1.83GHz - 2GB - Windows XP SP3 |
#293
|
|||
|
|||
Why do you still use Windows XP?
"BillW50" wrote in :
Anyway being new to RAMDisks, I was done with coding and powered off the computer without thinking to save my work to a floppy. Instantly I thought oh crap! Powered back on in less than a second and checked the RAMDisk and everything was still there fully intact. ;-) So I started testing how long the power could be gone before corruption would occur. And it would still hold data for up to 4 to 7 seconds later. 5 to 7 seconds was a toss up and it could go either way. I wished I investigated further to learn how long those RAM chips could retain data without a refresh. As I would guess a second or so there could have been enough power floating around to keep the refresh going for a bit. But even that doesn't explain up to 7 seconds later and everything could still be intact. That's useful. I don't know how it works but anything that persists even for a moment is good. Our eyes work on persistence of vision for continuity, and as far as I know, brains don't do caching the way computers do. Maybe there's something to temporal persistence that can avoid the need. Just rambling here, but something about this seems to imply some powerful discoveries in any number of useful contexts. One obvious known one is the use of flywheels in engines to overcome the need to cache extra fuel for bursts of acceleration, but I bet there are a lot of non-obvious ones to come. I guess the whole RAM refresh thing is based on this, otherwise it would be all refresh and not a lot of other access! |
#294
|
|||
|
|||
Why do you still use Windows XP?
In message , BillW50
writes: [] I am not sure if there is enough demand for Windows Embedded anyway? Since the appearance of netbooks, tablets, and such... it seems to make more sense just running the stock Windows anyway, don't you think? Oh, I think systems that were (more or less) instant on, could not be corrupted, and looked like normal Windows, would have a not-small market. -- J. P. Gilliver. UMRA: 1960/1985 MB++G.5AL-IS-P--Ch++(p)Ar@T0H+Sh0!:`)DNAf I believe the cake has got to be sliced up to help those who are needy and you've got to keep someone there who's going to make the cake. Here we always destroy the people who make the cake. - Michael Caine (MM), RT, 7-13 Nov 2009. |
#295
|
|||
|
|||
Why do you still use Windows XP?
|
#296
|
|||
|
|||
Why do you still use Windows XP?
On Tue, 20 Mar 2012 08:36:46 +0000, "J. P. Gilliver (John)"
wrote: In message , BillW50 writes: [] I am not sure if there is enough demand for Windows Embedded anyway? Since the appearance of netbooks, tablets, and such... it seems to make more sense just running the stock Windows anyway, don't you think? Oh, I think systems that were (more or less) instant on, could not be corrupted, and looked like normal Windows, would have a not-small market. On the other hand, they'd be quickly exploited. |
#297
|
|||
|
|||
Why do you still use Windows XP?
|
#298
|
|||
|
|||
Why do you still use Windows XP?
On Tue, 20 Mar 2012 10:04:42 -0400, Zaphod Beeblebrox
wrote: In article , says... On Tue, 20 Mar 2012 08:36:46 +0000, "J. P. Gilliver (John)" wrote: In message , BillW50 writes: [] I am not sure if there is enough demand for Windows Embedded anyway? Since the appearance of netbooks, tablets, and such... it seems to make more sense just running the stock Windows anyway, don't you think? Oh, I think systems that were (more or less) instant on, could not be corrupted, and looked like normal Windows, would have a not-small market. On the other hand, they'd be quickly exploited. What do you mean? I mean in the usual sense. |
#299
|
|||
|
|||
Why do you still use Windows XP?
|
Thread Tools | |
Display Modes | |
|
|