This is not true. TC uses k for 1024 and M for 1024*1024. So I bet it uses G for 1024*1024*1024.JackFoo wrote:What TC does is the first time divides by 1024 (bytes->kbytes) then uses 1000 as the next step ...
proper (standardized) file size symbols?
Moderators: white, Hacker, petermad, Stefan2
Hmm, didn't know that, thanks for the info.I am quite sure that Christian heard of the SI standard because almost all Swiss high schools teach it.
BTW why would they teach you standards in school? Seems like a waste of time... schools should teach you useful info that will drive you to expand your mind (especially in computers).
My bad again (damn twice in a post), just tried it again and you're correct about M (miscalculated it the first time); This however invokes the question about using lower k and caps for G and M...?This is not true. TC uses k for 1024 and M for 1024*1024. So I bet it uses G for 1024*1024*1024
Cheers.
I imagine it is because 'K' is the SI-symbol for Kelvin, the base unit of thermodynamic temperature, or the other way round because 'g' stands for gram (but oddly enough 'kg' is the base unit of mass) and 'm' stands for meter, the base unit of length. I really recommend you to get acquainted with the SI standard because it is not that irrelevant as you think.JackFoo wrote:This however invokes the question about using lower k and caps for G and M...?
yes thats most likely the reason :)jb wrote:I imagine it is because 'K' is the SI-symbol for Kelvin, the base unit of thermodynamic temperature, or the other way round because 'g' stands for gram (but oddly enough 'kg' is the base unit of mass) and 'm' stands for meter, the base unit of length. I really recommend you to get acquainted with the SI standard because it is not that irrelevant as you think.JackFoo wrote:This however invokes the question about using lower k and caps for G and M...?
i suppose IEC resides in Switzerland, or at least their web site :)
and afaik the SI standard is taught in basic school in most of european countries (at least those i know)
standards are not created out of whim of some scientists, they are needed to avoid confusion, inaccuracies, misuse, etc.
Grave
*eat people, not animals*
*eat people, not animals*
I have nothing against standards, they are important tools for seamless communication; Sometimes however standards become the goal rather than a tool. I know what kg stands for, it wouldn't bother me if someone wrote it as KG because context would clarify the meaning.
But we're getting OT, I think TC's convention of symbols is quite adequate (it wouldn't heart if it was also explained in the help; I don't currently have access to TC's help but I'll check it out)
Cheers.
But we're getting OT, I think TC's convention of symbols is quite adequate (it wouldn't heart if it was also explained in the help; I don't currently have access to TC's help but I'll check it out)
Cheers.
I agree with the opinion that ambiguity is a bad thing. And it would be more convenient and easy to have in mind 1000 seing 1KB and 1000000 seing 1MB. Because it is easy to calculate in mind, and that's what we do in everyday life with grams, meters, etc.
These kibi, mibi, gibi, ... sound quite funny for Russian/Ukrainian ear...
Why at all we need 1024 multiples? Yes, all chips are operating with two levels of voltage representing bit values. So what?
Nevertheless, it would be quite hard to re-learn the whole world (well, quite big part of the world).
These kibi, mibi, gibi, ... sound quite funny for Russian/Ukrainian ear...
Why at all we need 1024 multiples? Yes, all chips are operating with two levels of voltage representing bit values. So what?
Nevertheless, it would be quite hard to re-learn the whole world (well, quite big part of the world
).
Last edited by Valentino on 2003-06-03, 00:59 UTC, edited 1 time in total.
Ok I've checked TC's help:
Note the misspelled kbytes (kybes - I think this isn't one of the weird names) and depending (depeding); BTW kibi, mibi sound completely silly to me as well.
It seems docs only specify conversion from b->kb, but not M or G IMO those should be added.
Now, this is an interesting passage, so TC calculates the real space a file takes (i.e. cluster size etc.) so perhaps it does use M, G for x1000... since help only specifies 1024b=1kb and doesn't say anything about M or G it seems likely to me that my second mistake wasn't a mistake at all but rather some cluster space taken into account. Which seems logical. Perhaps Christian can shed some light on this.
Cheers.
Code: Select all
Size display:
Determines how the file size is shown in the main file lists. You can choose between
bytes The size is shown in bytes, as in previous versions of Total Commander (default)
kbytes Size is shown in kybes. 1 kbyte = 1024 bytes!
dynamic (x k/M/G) depeding on the file size, it is displayed in bytes, kbytes, MBytes or even GBytes
dynamic (x.x k/M/G) same as above, but with one digit after the comma to get a more exact size
It seems docs only specify conversion from b->kb, but not M or G IMO those should be added.
Code: Select all
Calculate occupied space
Calculates the space occupied by the marked files and their complete subdirectories. If no file is selected, the space occupied by the file or directory under the cursor bar is calculated. Additionally, the file size is rounded to the cluster size of the source and destination directory, to get real space required. The space for the directory information itself is not considered.
Cheers.
JackFoo,
small [k] stands for kilo (x1000) but upper case K stands for x1024 the same goes for M and G
That unfortunately doesn't work with M... if it did it would make our life much easier.
Roman
small [k] stands for kilo (x1000) but upper case K stands for x1024 the same goes for M and G
That unfortunately doesn't work with M... if it did it would make our life much easier.
Roman
Mal angenommen, du drückst Strg+F, wählst die FTP-Verbindung (mit gespeichertem Passwort), klickst aber nicht auf Verbinden, sondern fällst tot um.
lol yes they sound silly and hilarious at first (kiwi comes in mind :))JackFoo wrote:BTW kibi, mibi sound completely silly to me as well.
but they kinda grow on you when you start using them :)
(esp. when you can say "i have 200 gibs of data", sounds pretty gory :D)
still having it at least optional wouldnt be bad thing
Grave
*eat people, not animals*
*eat people, not animals*
Yep, true already figured it out (see later posts); It seems TC uses the following conversions:Hacker wrote: That unfortunately doesn't work with M... if it did it would make our life much easier.
bytes -> kilo (x1024)
kilo -> mega (x1000)
mega -> giga (x1000)
The last 2 conversions aren't certain but I'm pretty sure they are correct; there was some argument and JB pointed out that all conversions are done using x1024 but I think this a mistake stemming from the fact that TC shows the actual physical space taken (cluster size taken into account).
BTW I have never heard anyone say gibi or kibi or any other of the names and I can safely guess that I never will; I would probably laugh at the first person to utter these words... (just as the people I know would have done). Actually it's probably one of the downfalls of this standard, perhaps it's adoption would have been better if they chose normal names.
Cheers.
.
AFAIK, cluster size is taken into account only in Ctrl+L feature (cm_GetFileSpace).JackFoo wrote:the fact that TC shows the actual physical space taken (cluster size taken into account)
About K->M conversion. I made very simple test: the file 4167552 bytes long is displayed as "3.9M" so M=1024 K, because 4167552/1024/1024=3,974487304688.
And as you can also see (someone already mentioned this somewhere) TC doesn't round sizes but just truncates them. Not correct, IMhO because rounding will display more accurate values.
No, it's obvious that kibble stands for kiblo-byte...VadiMGP wrote:2JohnFredC
Why you confused? It is obviously bles containing 1024 binary bytes and several bits!
Seems that that food is for advanced PC-competent dogs...I must pronounce them several times every morning in front of my mirror to get used to them...Grave wrote:but they kinda grow on you when you start using them
I must say it again: This is wrong. M is 1024*1024 in TC. Furthermore such an irregular scheme as you insist on would not be like TC/Christian at all. We are just talking about a notation. So cluster size and stuff like that is completely irrelevant here.JackFoo wrote:It seems TC uses the following conversions:
bytes -> kilo (x1024)
kilo -> mega (x1000)
mega -> giga (x1000)
The last 2 conversions aren't certain but I'm pretty sure they are correct; there was some argument and JB pointed out that all conversions are done using x1024 but I think this a mistake stemming from the fact that TC shows the actual physical space taken (cluster size taken into account).
Although not explicitly expressed in the post somewhere (where and when was it again? In the Internet 2 years ago?Valentino wrote:About K->M conversion. I made very simple test: the file 4167552 bytes long is displayed as "3.9M" so M=1024 K, because 4167552/1024/1024=3,974487304688.
And as you can also see (someone already mentioned this somewhere) TC doesn't round sizes but just truncates them. Not correct, IMhO because rounding will display more accurate values.
) someone agrees that rounding is preferable because it is a FACT (not just an opinion) that rounding is more accurate than truncating generally.