What is the best results achieved in compressing the Random Million digit 
bin file?  1 byte? 2 bytes? more?

Paul 

On Jun 24, 10:55=A0am, "Paul" <p...@tretbase.com> wrote:
> What is the best results achieved in compressing the Random Million digit
> bin file? =A01 byte? 2 bytes? more?
>
> Paul

With or without including the size of the decompressor, filenames, and
any related metadata?

Tom

"Tom St Denis" <tom@iahu.ca> wrote in message 
news:6bb1f580-8f91-4145-b9a3-08008c78b3a0@e20g2000vbc.googlegroups.com...
> On Jun 24, 10:55 am, "Paul" <p...@tretbase.com> wrote:
>> What is the best results achieved in compressing the Random Million digit
>> bin file?  1 byte? 2 bytes? more?
>>
>> Paul
>
> With or without including the size of the decompressor, filenames, and
> any related metadata?
>
> Tom

Yes Tom, I just want to know the highest level of compression accomplished 
in bytes for just the data itself, excluding all else.

Paul 

On Jun 24, 10:01=A0am, "Paul" <p...@tretbase.com> wrote:
> "Tom St Denis" <t...@iahu.ca> wrote in messagenews:6bb1f580-8f91-4145-b9a=
3-08008c78b3a0@e20g2000vbc.googlegroups.com...
>
> > On Jun 24, 10:55 am, "Paul" <p...@tretbase.com> wrote:
> >> What is the best results achieved in compressing the Random Million di=
git
> >> bin file? =A01 byte? 2 bytes? more?
>
> >> Paul
>
> > With or without including the size of the decompressor, filenames, and
> > any related metadata?
>
> > Tom
>
> Yes Tom, I just want to know the highest level of compression accomplishe=
d
> in bytes for just the data itself, excluding all else.
>
> Paul

"Yes" what? You failed to answer the question.

This is important because I can easily reduce the million random
digits file
to exactly 0 (yes, zero) "data" bytes if you don't count all that
other stuff.

"Tom St Denis" <tom@iahu.ca> wrote in message 
news:6bb1f580-8f91-4145-b9a3-08008c78b3a0@e20g2000vbc.googlegroups.com...
> On Jun 24, 10:55 am, "Paul" <p...@tretbase.com> wrote:
>> What is the best results achieved in compressing the Random Million digit
>> bin file?  1 byte? 2 bytes? more?
>>
>> Paul
>
> With or without including the size of the decompressor, filenames, and
> any related metadata?
>
> Tom

Without including the size of the decompressor, filenames and any related 
metadata.

Paul 

In article <bhu0m.49540$lB7.12960@newsfe19.iad>,
Paul <paul@tretbase.com> wrote:
>
>
>"Tom St Denis" <tom@iahu.ca> wrote in message 
>news:6bb1f580-8f91-4145-b9a3-08008c78b3a0@e20g2000vbc.googlegroups.com...
>> On Jun 24, 10:55 am, "Paul" <p...@tretbase.com> wrote:
>>> What is the best results achieved in compressing the Random Million digit
>>> bin file?  1 byte? 2 bytes? more?
>>>
>>> Paul
>>
>> With or without including the size of the decompressor, filenames, and
>> any related metadata?
>>
>> Tom
>
>Without including the size of the decompressor, filenames and any related 
>metadata.

I can compress it down to one byte (all other files get expanded by one byte,
which isn't bad).  The decompressor is pretty big, however.

Alan
-- 
Defendit numerus

Paul wrote:

>>> What is the best results achieved in compressing the Random Million 
>>> digit
>>> bin file?  1 byte? 2 bytes? more?
>>>
>>> Paul
>>
>> With or without including the size of the decompressor, filenames, and
>> any related metadata?
>>
>> Tom
> 
> Without including the size of the decompressor, filenames and any 
> related metadata.

0 bytes. Literally: ZERO. I can write a decompressor for that easily.

Thus, I afraid, your question is rather meaningless.

Greetings,
	Thomas

"Thomas Richter" <thor@math.tu-berlin.de> wrote in message 
news:h1tt61$g9h$1@infosun2.rus.uni-stuttgart.de...
> Paul wrote:
>
>>>> What is the best results achieved in compressing the Random Million 
>>>> digit
>>>> bin file?  1 byte? 2 bytes? more?
>>>>
>>>> Paul
>>>
>>> With or without including the size of the decompressor, filenames, and
>>> any related metadata?
>>>
>>> Tom
>>
>> Without including the size of the decompressor, filenames and any related 
>> metadata.
>
> 0 bytes. Literally: ZERO. I can write a decompressor for that easily.
>
> Thus, I afraid, your question is rather meaningless.
>
> Greetings,
> Thomas

yes, I guess I don't know how to explain what I'm looking for.  Oh well.  I 
won't reveal anything then until I get someone who can understand what I'm 
looking for.

Paul 

"Paul" <paul@tretbase.com> wrote in message 
news:x7r0m.40738$QS.15405@newsfe18.iad...
> What is the best results achieved in compressing the Random Million digit 
> bin file?  1 byte? 2 bytes? more?
>
> Paul

I'll start another thread as people couldn't understand what I was asking.

Paul 

Paul wrote:

>>> Without including the size of the decompressor, filenames and any 
>>> related metadata.
>>
>> 0 bytes. Literally: ZERO. I can write a decompressor for that easily.
>>
>> Thus, I afraid, your question is rather meaningless.
>>
>> Greetings,
>> Thomas
> 
> yes, I guess I don't know how to explain what I'm looking for.  Oh 
> well.  I won't reveal anything then until I get someone who can 
> understand what I'm looking for.

Well, I afraid you first need to understand the question. The problem is 
that you cannot measure the efficiency of a compression scheme on *one 
specific* source. You can only measure it on a *class of sources* of 
which the one you picked is a typical one.

For example, a more meaningful question would be:

a) what is the minimal size of the compressed file and the decompressor 
(two files only, file names do not matter): This would be interpreting 
the compressor in the sense of the Kolmogorov complexity (one method).

b) What is the minimal size a gaussian i.i.d. source with uniform 
probability density can be compressed to? (This would be a model for the 
given file). The answer here is known: The size of the file itself (on 
average).

Thanks,
	Thomas

On Jun 24, 10:55=A0am, "Paul" <p...@tretbase.com> wrote:
> What is the best results achieved in compressing the Random Million digit
> bin file? =A01 byte? 2 bytes? more?

Basicly, you can prove your compressor/decompressor works if the all
of the resulting compressed_files, metadata and decompressor can be
tarballed into a single file that is 50 bytes or smaller than the
binary version of the 'Million-Digits-File' that any of us can
download and test on our own trusted systems.

Your claims are meaningless, they are just words.  Your computer code
running on your system is untrusted because of possible cheats and/or
possible mistakes on your part when testing.  Your witnesses can not
be trusted until it is clear they have not been brided or in-fact do
not know how to properly test such a compression program.

If you want to prove it, you have to show it.

And the easy way is simple.  Send us your decompressor, send us your
compressed version of the 'Million-Digits-File' and tarballed together
would be best.

A Tarball that is 50 bytes or more smaller than the binary version of
the 'Million-Digits-File' is what will really make us sit up and
notice.

And ***WE*** will run it on hardware that we can trust.  Hardware with
no internet connections, hardware with no hidden version of the
'Million-Digits-File', hardware under our own personal control.  If
you can then pass all of the above, I will personally kiss your shoe.
Hech, I will post a video of it on YouTube if you want.

But the Tarball of *ALL* the needed files must be 50 bytes or more
smaller than the binary version of the 'Million-Digits-File' if you
want my beautiful lips on your smelly shoes.

PS. It's been 15 years of baiting people who make claims like you, and
I have not lost a single bet yet.

On Jun 25, 8:08=A0am, Earl_Colby_Pottinger
<earlcolby.pottin...@sympatico.ca> wrote:
> On Jun 24, 10:55=A0am, "Paul" <p...@tretbase.com> wrote:
>
> > What is the best results achieved in compressing the Random Million dig=
it
> > bin file? =A01 byte? 2 bytes? more?
>
> Basicly, you can prove your compressor/decompressor works if the all
> of the resulting compressed_files, metadata and decompressor can be
> tarballed into a single file that is 50 bytes or smaller than the
> binary version of the 'Million-Digits-File' that any of us can
> download and test on our own trusted systems.
>
> Your claims are meaningless, they are just words. =A0Your computer code
> running on your system is untrusted because of possible cheats and/or
> possible mistakes on your part when testing. =A0Your witnesses can not
> be trusted until it is clear they have not been brided or in-fact do
> not know how to properly test such a compression program.
>
> If you want to prove it, you have to show it.
>
> And the easy way is simple. =A0Send us your decompressor, send us your
> compressed version of the 'Million-Digits-File' and tarballed together
> would be best.
>
> A Tarball that is 50 bytes or more smaller than the binary version of
> the 'Million-Digits-File' is what will really make us sit up and
> notice.
>
> And ***WE*** will run it on hardware that we can trust. =A0Hardware with
> no internet connections, hardware with no hidden version of the
> 'Million-Digits-File', hardware under our own personal control. =A0If
> you can then pass all of the above, I will personally kiss your shoe.
> Hech, I will post a video of it on YouTube if you want.
>
> But the Tarball of *ALL* the needed files must be 50 bytes or more
> smaller than the binary version of the 'Million-Digits-File' if you
> want my beautiful lips on your smelly shoes.
>
> PS. It's been 15 years of baiting people who make claims like you, and
> I have not lost a single bet yet.

Like that,

I determined to get your kiss too.


Regards,
Fibonacci.
-p.s. joke