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 | Rate Thread | Display Modes |
#61
|
|||
|
|||
32 GB memory stick
In message , Paul
writes: [] To make USB3 pen drives, they have a problem fitting enough Flash channels in parallel, to achieve the same rates we see from SSD drives. The pen drives I've seen for sale, are already pretty fat and unwieldy. Paul What does the "D" stand for in "SSD drives"? -- J. P. Gilliver. UMRA: 1960/1985 MB++G.5AL-IS-P--Ch++(p)Ar@T0H+Sh0!:`)DNAf After I'm dead I'd rather have people ask why I have no monument than why I have one. -Cato the Elder, statesman, soldier, and writer (234-149 BCE) |
Ads |
#62
|
|||
|
|||
32 GB memory stick
On Thu, 10 Nov 2011 21:50:40 +0000, "J. P. Gilliver (John)"
wrote: In message , Paul writes: [] To make USB3 pen drives, they have a problem fitting enough Flash channels in parallel, to achieve the same rates we see from SSD drives. The pen drives I've seen for sale, are already pretty fat and unwieldy. Paul What does the "D" stand for in "SSD drives"? It's similar to the "C" in "NIC card". -- Char Jackson |
#63
|
|||
|
|||
32 GB memory stick
On 10/11/2011 21:32, R. C. White wrote:
Hi, Peter. "Peter Jason" wrote in message ... SNIP Ok, sorry, I was going by the USB3 spec. I see now that the bottleneck isn't the USB3 port but rather the device itself. If the packaging suggests a read rate of about 135MB/sec and you're only getting 137-152 Mb/sec, then something is still wrong. (Unless you're still confusing Mb with MB? Megabits versus Megabytes) Now I'm confused. I need a rest. I am not a techie, but...a non-techie refresher for anyone confused by MB and Mb: It takes 8 bits to make a byte, as most computer addicts know. But to SEND a byte online, or even between disks or other devices within one computer, the sender has to tell the receiver WHERE in the string of bits a byte begins. So, before each batch of 8 bits it inserts a START bit, then it ends each byte with a STOP bit, making a 10-bit transmission for each byte. To send 1,000 bytes, the transmitter must send 10,000 bits. The abbreviation for "bits" is "b", the lowercase "b"; the uppercase "B" is used for "bytes". So 135 MB/sec is 10 times as fast as 135 Mb/sec. Simple - but oh, so easy to overlook the often-subtle difference between "b" and "B" when reading along. :^{ RC -- Add a parity bit and things get even bigger. Ed |
#64
|
|||
|
|||
32 GB memory stick
Hi, John.
SSD = Solid State Device - or Solid State Drive NIC = Network Interface Card I think. Char can correct me if I'm wrong. ;^} RC -- R. C. White, CPA San Marcos, TX Microsoft Windows MVP (2002-2010) Windows Live Mail 2011 (Build 15.4.3538.0513) in Win7 Ultimate x64 SP1 What does the "D" stand for in "SSD drives"? -- J. P. Gilliver. UMRA: 1960/1985 MB++G.5AL-IS-P--Ch++(p)Ar@T0H+Sh0!:`)DNAf |
#65
|
|||
|
|||
32 GB memory stick
On Thu, 10 Nov 2011 15:32:01 -0600, R. C. White wrote:
Hi, Peter. "Peter Jason" wrote in message ... SNIP Ok, sorry, I was going by the USB3 spec. I see now that the bottleneck isn't the USB3 port but rather the device itself. If the packaging suggests a read rate of about 135MB/sec and you're only getting 137-152 Mb/sec, then something is still wrong. (Unless you're still confusing Mb with MB? Megabits versus Megabytes) Now I'm confused. I need a rest. I am not a techie, but...a non-techie refresher for anyone confused by MB and Mb: It takes 8 bits to make a byte, as most computer addicts know. But to SEND a byte online, or even between disks or other devices within one computer, the sender has to tell the receiver WHERE in the string of bits a byte begins. So, before each batch of 8 bits it inserts a START bit, then it ends each byte with a STOP bit, making a 10-bit transmission for each byte. To send 1,000 bytes, the transmitter must send 10,000 bits. The abbreviation for "bits" is "b", the lowercase "b"; the uppercase "B" is used for "bytes". So 135 MB/sec is 10 times as fast as 135 Mb/sec. Simple - but oh, so easy to overlook the often-subtle difference between "b" and "B" when reading along. :^{ RC The above is true for some old serial protocols, but not used a lot today. All protocols have some overhead, but not as much as described above. And the blocks are not usually eight bits any more, but more like 128 or 256 bits, in fact, often 128 or 256 data bits plus some protocol bits. Most transmission blocks have a few byte sat the beginning giving such things as the protocol in use, the category or type of the block, the ID or sequence number of the block, the source and destination IDs, and in some cases a byte count. The sequence number is especially useful when there are multiple data streams or when there are retransmissions. The blocks usually have, most likely at the end, a check sum or a set of check digits designed to detect a corrupted block. If the block isn't too corrupted, these bits may even allow reconstructing the block's data, but if that can't be done, the receiver requests a retransmission. Nowhere near as simple as above. And hardly new. I first worked with that stuff around 1985; I hardly think it was new then. The above is the best I can do for now, since 1985 is also pretty much when I *last* worked with that stuff :-) -- Gene E. Bloch (Stumbling Bloch) |
#66
|
|||
|
|||
32 GB memory stick
On Thu, 10 Nov 2011 15:32:01 -0600, "R. C. White"
wrote: I am not a techie, but...a non-techie refresher for anyone confused by MB and Mb: It takes 8 bits to make a byte, as most computer addicts know. But to SEND a byte online, or even between disks or other devices within one computer, the sender has to tell the receiver WHERE in the string of bits a byte begins. So, before each batch of 8 bits it inserts a START bit, then it ends each byte with a STOP bit, making a 10-bit transmission for each byte. To send 1,000 bytes, the transmitter must send 10,000 bits. The abbreviation for "bits" is "b", the lowercase "b"; the uppercase "B" is used for "bytes". So 135 MB/sec is 10 times as fast as 135 Mb/sec. Simple - but oh, so easy to overlook the often-subtle difference between "b" and "B" when reading along. :^{ Thanks, R.C. I suppose I should have explained the 'b versus B' thing rather than assume the OP knew what I was talking about. Thanks for doing it. Regarding the per-byte start, stop, and parity bits, my understanding is that those things are pretty much a quaint bit of history that disappeared along with dial-up modems. These days most of us (right?) have replaced our modems with Ethernet connections that carry IP traffic. IP data packets have headers and payload, where the payload can be TCP, for example, which itself has headers and payload, and so on. The whole packet will have a checksum rather than attempting to verify the integrity of each individual byte. Then on the local (LAN) network, everything is stuffed into Ethernet frames, and they too have headers and payload. Even with all of that, the net result is that there is less overhead than there was with dial-up. That's probably simplified to the point that much of it is wrong, but I'm going to quit while I'm behind. -- Char Jackson |
#67
|
|||
|
|||
32 GB memory stick
On Thu, 10 Nov 2011 18:06:06 -0600, Char Jackson wrote:
On Thu, 10 Nov 2011 15:32:01 -0600, "R. C. White" wrote: I am not a techie, but...a non-techie refresher for anyone confused by MB and Mb: It takes 8 bits to make a byte, as most computer addicts know. But to SEND a byte online, or even between disks or other devices within one computer, the sender has to tell the receiver WHERE in the string of bits a byte begins. So, before each batch of 8 bits it inserts a START bit, then it ends each byte with a STOP bit, making a 10-bit transmission for each byte. To send 1,000 bytes, the transmitter must send 10,000 bits. The abbreviation for "bits" is "b", the lowercase "b"; the uppercase "B" is used for "bytes". So 135 MB/sec is 10 times as fast as 135 Mb/sec. Simple - but oh, so easy to overlook the often-subtle difference between "b" and "B" when reading along. :^{ Thanks, R.C. I suppose I should have explained the 'b versus B' thing rather than assume the OP knew what I was talking about. Thanks for doing it. Regarding the per-byte start, stop, and parity bits, my understanding is that those things are pretty much a quaint bit of history that disappeared along with dial-up modems. These days most of us (right?) have replaced our modems with Ethernet connections that carry IP traffic. IP data packets have headers and payload, where the payload can be TCP, for example, which itself has headers and payload, and so on. The whole packet will have a checksum rather than attempting to verify the integrity of each individual byte. Then on the local (LAN) network, everything is stuffed into Ethernet frames, and they too have headers and payload. Even with all of that, the net result is that there is less overhead than there was with dial-up. That's probably simplified to the point that much of it is wrong, but I'm going to quit while I'm behind. I think between us, with our different approaches, we have a pretty good summary of transmission protocols. And what we lack in details we make up for with the cogency of our explanations :-) I especially like that you pointed out the Matroshka nature of these protocols (i.e., like nesting Russian dolls). Wish I'd thought of that :-) -- Gene E. Bloch (Stumbling Bloch) |
#68
|
|||
|
|||
32 GB memory stick
On Thu, 10 Nov 2011 17:05:19 -0600, "R. C. White"
wrote: Hi, John. SSD = Solid State Device - or Solid State Drive NIC = Network Interface Card I think. Char can correct me if I'm wrong. ;^} I would never do that. ;-) -- Char Jackson |
#69
|
|||
|
|||
32 GB memory stick
On Thu, 10 Nov 2011 16:31:09 -0800, "Gene E. Bloch"
wrote: I think between us, with our different approaches, we have a pretty good summary of transmission protocols. And what we lack in details we make up for with the cogency of our explanations :-) I especially like that you pointed out the Matroshka nature of these protocols (i.e., like nesting Russian dolls). Wish I'd thought of that :-) I ended up mostly rambling, but the tag team approach is sometimes best. Thanks. -- Char Jackson |
#70
|
|||
|
|||
32 GB memory stick
In message , Char Jackson
writes: On Thu, 10 Nov 2011 21:50:40 +0000, "J. P. Gilliver (John)" wrote: In message , Paul writes: [] To make USB3 pen drives, they have a problem fitting enough Flash channels in parallel, to achieve the same rates we see from SSD drives. The pen drives I've seen for sale, are already pretty fat and unwieldy. Paul What does the "D" stand for in "SSD drives"? It's similar to the "C" in "NIC card". That's what I thought. (Another newsgroup I take refers to this as "PNS syndrome", where PN is "PIN number".) But I'll let him off this once because it _could_ mean "device" - though usually doesn't (-:. -- J. P. Gilliver. UMRA: 1960/1985 MB++G.5AL-IS-P--Ch++(p)Ar@T0H+Sh0!:`)DNAf After I'm dead I'd rather have people ask why I have no monument than why I have one. -Cato the Elder, statesman, soldier, and writer (234-149 BCE) |
#71
|
|||
|
|||
32 GB memory stick
In message , Gene E. Bloch
writes: On Thu, 10 Nov 2011 15:32:01 -0600, R. C. White wrote: [] It takes 8 bits to make a byte, as most computer addicts know. But to SEND a byte online, or even between disks or other devices within one computer, the sender has to tell the receiver WHERE in the string of bits a byte begins. So, before each batch of 8 bits it inserts a START bit, then it ends each byte with a STOP bit, making a 10-bit transmission for each byte. To send 1,000 bytes, the transmitter must send 10,000 bits. Just adding a start and/or stop bit doesn't of itself help much with the framing: to achieve that you have to add an _odd_ number of half bits. Otherwise, depending on the data, you can go for a long time without lining up properly. [] The above is true for some old serial protocols, but not used a lot today. [] Indeed - modern protocols are different, and nested, as others have said (and sometimes include sections that are either not required but there for historical reasons, or are at least duplicated in the several layers). But we still send bits serially - there's only one line, though with multiphase encoding this can send several bits at once, so it would be more accurate to say we send symbols serially - so there _is_ still a need for _some_ method of maintaining framing. (A point often infuriatingly omitted in papers comparing MFM, PSK, AM, and all the others.) -- J. P. Gilliver. UMRA: 1960/1985 MB++G.5AL-IS-P--Ch++(p)Ar@T0H+Sh0!:`)DNAf After I'm dead I'd rather have people ask why I have no monument than why I have one. -Cato the Elder, statesman, soldier, and writer (234-149 BCE) |
#72
|
|||
|
|||
32 GB memory stick
On Fri, 11 Nov 2011 08:08:50 +0000, "J. P. Gilliver (John)"
wrote: That's what I thought. (Another newsgroup I take refers to this as "PNS syndrome", where PN is "PIN number".) But I'll let him off this once because it _could_ mean "device" - though usually doesn't (-:. -- It is possible to have more than one PIN so PIN number could make sense. Steve -- Neural network software applications, help and support. Neural Network Software. www.npsl1.com EasyNN-plus. Neural Networks plus. www.easynn.com SwingNN. Forecast with Neural Networks. www.swingnn.com JustNN. Just Neural Networks. www.justnn.com |
#73
|
|||
|
|||
32 GB memory stick
On Fri, 11 Nov 2011 10:05:20 +0000, Stephen Wolstenholme wrote:
On Fri, 11 Nov 2011 08:08:50 +0000, "J. P. Gilliver (John)" wrote: That's what I thought. (Another newsgroup I take refers to this as "PNS syndrome", where PN is "PIN number".) But I'll let him off this once because it _could_ mean "device" - though usually doesn't (-:. -- It is possible to have more than one PIN so PIN number could make sense. Steve Now you're toying with us :-) And I have to admit I never thought of it! In the area where I live, a major thoroughfare is El Camino Real, and people often refer to it as "the El Camino". Now I ask for your help in coming up with a useful parsing of that :-) -- Gene E. Bloch (Stumbling Bloch) |
#74
|
|||
|
|||
32 GB memory stick
On Fri, 11 Nov 2011 08:16:32 +0000, J. P. Gilliver (John) wrote:
In message , Gene E. Bloch writes: On Thu, 10 Nov 2011 15:32:01 -0600, R. C. White wrote: [] It takes 8 bits to make a byte, as most computer addicts know. But to SEND a byte online, or even between disks or other devices within one computer, the sender has to tell the receiver WHERE in the string of bits a byte begins. So, before each batch of 8 bits it inserts a START bit, then it ends each byte with a STOP bit, making a 10-bit transmission for each byte. To send 1,000 bytes, the transmitter must send 10,000 bits. Just adding a start and/or stop bit doesn't of itself help much with the framing: to achieve that you have to add an _odd_ number of half bits. Otherwise, depending on the data, you can go for a long time without lining up properly. [] The above is true for some old serial protocols, but not used a lot today. [] Indeed - modern protocols are different, and nested, as others have said (and sometimes include sections that are either not required but there for historical reasons, or are at least duplicated in the several layers). But we still send bits serially - there's only one line, though with multiphase encoding this can send several bits at once, so it would be more accurate to say we send symbols serially - so there _is_ still a need for _some_ method of maintaining framing. (A point often infuriatingly omitted in papers comparing MFM, PSK, AM, and all the others.) Of course, with the protocols we seem to be thinking of here, missing the correct start of a block would make the block uninterpretable. The receiver couldn't even detect a checksum error, because it wouldn't know where the checksum is. Hence the receiver would have to ignore the block, and hopefully, the transmitter, lacking the moral equivalent of an ACK or a NAK, would eventually retransmit the block. Please pardon the anthropomorphic writing - I'm just being lazy :-) -- Gene E. Bloch (Stumbling Bloch) |
#75
|
|||
|
|||
32 GB memory stick
On 11/11/2011 18:51, Gene E. Bloch wrote:
On Fri, 11 Nov 2011 10:05:20 +0000, Stephen Wolstenholme wrote: On Fri, 11 Nov 2011 08:08:50 +0000, "J. P. Gilliver (John)" wrote: That's what I thought. (Another newsgroup I take refers to this as "PNS syndrome", where PN is "PIN number".) But I'll let him off this once because it _could_ mean "device" - though usually doesn't (-:. -- It is possible to have more than one PIN so PIN number could make sense. Steve Now you're toying with us :-) And I have to admit I never thought of it! In the area where I live, a major thoroughfare is El Camino Real, and people often refer to it as "the El Camino". Now I ask for your help in coming up with a useful parsing of that :-) I see completely what you mean. Here's another in the same category. Is Las Vegas a good town? Is Buenos Aires near Rio? Are the airs good there? Ed |
Thread Tools | |
Display Modes | Rate This Thread |
|
|