Which encryption algorithm should I use setting up Truecrypt volume?

Our family computer is a laptop, and since it would be very easy and
desireable to steal if someone broke into our house, I figured I should
get some encryption to protect our data. So I downloaded and installed
Truecrypt. But during the setup of my first volume, it's asking about
which encryption algorithm I want to use. I don't really know what
would be best, so I figured I should ask people who know. Thanks for
your help...

rinmanb70

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

"rinmanb70" <> wrote in
news: oups.com:

> Our family computer is a laptop, and since it would be very easy
> and desireable to steal if someone broke into our house, I
> figured I should get some encryption to protect our data. So I
> downloaded and installed Truecrypt. But during the setup of my
> first volume, it's asking about which encryption algorithm I
> want to use. I don't really know what would be best, so I
> figured I should ask people who know. Thanks for your help...

Just leave the default one. There are no weak encryption
algorithms implemented in TC so you will be safe with any of them.

Regards,
cypher

-----BEGIN PGP SIGNATURE-----
Version: PGP 8.1

iQA/AwUBRJsRqSPnLg7nPH4AEQLsnQCbBgcFg8N8lfkSMHud2Gj4Mn L4fLsAoOSx
NekVBxk+rY2nL6Ijl/LuNcKd
=vFME
-----END PGP SIGNATURE-----

"rinmanb70" <> wrote:

> Our family computer is a laptop, and since it would be very easy and
> desireable to steal if someone broke into our house, I figured I
> should get some encryption to protect our data. So I downloaded and
> installed Truecrypt. But during the setup of my first volume, it's
> asking about which encryption algorithm I want to use. I don't
> really know what would be best, so I figured I should ask people who
> know. Thanks for your help...

They're all good, plenty good enough to thwart your common street thug.
The main difference is going to be in speed, which unless you're doing
some drive intense operations like real time video editing you're not
going to notice anyway.

George Orwell wrote:
> They're all good, plenty good enough to thwart your common street thug.
> The main difference is going to be in speed, which unless you're doing
> some drive intense operations like real time video editing you're not
> going to notice anyway.


Speed is important. Which one(s) are fastest?

rinmanb70 wrote:
> George Orwell wrote:
>> They're all good, plenty good enough to thwart your common street thug.
>> The main difference is going to be in speed, which unless you're doing
>> some drive intense operations like real time video editing you're not
>> going to notice anyway.
>
> Speed is important. Which one(s) are fastest?

Security is important as well. Which one(s) have been proven resistent
towards Saturation Attack?

Sebastian Gottschalk wrote:
>
> Security is important as well. Which one(s) have been proven resistent
> towards Saturation Attack?

I've been told that any of the Truecrypt encryption methods should
provide adequate security for my needs. I'm now asking about speed.

If you have a question of your own, post your own thread Sebastian.

rinmanb70 wrote:
> Sebastian Gottschalk wrote:
>> Security is important as well. Which one(s) have been proven resistent
>> towards Saturation Attack?
>
> I've been told that any of the Truecrypt encryption methods should
> provide adequate security for my needs.

True. However, some algorithms are well more analyzed and proven. That's
why Rijndael became AES even thought it's not the fastest.

> If you have a question of your own, post your own thread Sebastian.

It's a rhetoric question with the obvious answer: Blowfish has no prove
for resistence against saturation attack whereas all AES candidates have
one, especially the one for Rijndael is exceptionally elegant.

"rinmanb70" <> wrote:

>
> George Orwell wrote:
> > They're all good, plenty good enough to thwart your common street
> > thug. The main difference is going to be in speed, which unless
> > you're doing some drive intense operations like real time video
> > editing you're not going to notice anyway.
>
>
> Speed is important. Which one(s) are fastest?

First of all, if speed is your thing than you need to seriously
consider the container type. File containers will typically be slower
than entire encrypted partitions. A *lot* slower in some cases. This
decision alone will have a much bigger impact on throughput than
algorithm choice.

That said, speed comparisons are going to be system dependant and
arbitrary. Any flat comparison might be misleading. And I think the
differences are going to be so minute in the real world that they're
insignificant. I'll offer a specific opinion later on, but first some
examples of why they're arbitrary.

Key size shouldn't directly directly impact speed, but it may change
the algorithm's "rank" among its constituents from a 32 bit to a 64 bit
environment.

Your specific usage can also affect speed. Encryption almost always
takes more clock cycles than decryption. For some algorithms the
difference is more pronounced. If your activity is more "write
oriented" you might want to consider an algorithm with a
better encrypt/decrypt "ratio".

There's a lot of other little gotchas that might make one algorithm
perform better than another on your system, and with your usage. But if
you really *must* know which algorithms are considered "faster"...

Rijndael (AES) and Twofish are two of the best performers on a
"standardized" Pentium Pro, in a standardized implementation. Twofish
is slightly faster at encryption, but Rijndael takes the checkered in
key setup. I think overall, Rijndael would be the "winner" in a
standard usage scenario. CAST is not so far behind these two. I'd say
you'd be good with any of the three. On the other end, 3DES is probably
going to be one of the slowest.

FWIW, if you get down to the nitpicking "what if" of the thing, Twofish
is thought to be slightly more secure than Rijndael too. <grin>

Non scrivetemi wrote:

> Rijndael (AES) and Twofish are two of the best performers on a
> "standardized" Pentium Pro, in a standardized implementation. Twofish
> is slightly faster at encryption, but Rijndael takes the checkered in
> key setup. I think overall, Rijndael would be the "winner" in a
> standard usage scenario. CAST is not so far behind these two. I'd say
> you'd be good with any of the three. On the other end, 3DES is probably
> going to be one of the slowest.

Blowfish is usually noticeably faster than AES. However, it also has a
low security margin.

> FWIW, if you get down to the nitpicking "what if" of the thing, Twofish
> is thought to be slightly more secure than Rijndael too. <grin>

Sure? AFAICS some Chinese researchers have shown significant
improvements on cryptoanalysis of slightly simplified and/or modified
TwoFish.

Thanks for your help everybody.
I used the default, AES, and I'll see how it goes. If it seems too
slow then I'll try another alogorithm or consider using an encrypted
partition instead of an encrypted file.