Re: Eric Gisse doesn't know the basics of Relativity

On Aug 6, 4:57 am, "T.M. Sommers" <t...@xxxxxx> wrote:
Eric Gisse wrote:
On Aug 5, 9:16 am, "T.M. Sommers" <t...@xxxxxx> wrote:
gu...@xxxxxxxxxxx wrote:
On Aug 5, 11:12 am, "T.M. Sommers" <t...@xxxxxx> wrote:

Temperature is not invariant. It transforms as

T' = T * (1 - v^2/c^2)^(1/2)

Come on. Rest Mass is a Lorentz invariant scalar.

Never heard of invariant mass, M' = gamma M.

M remains invariant.

I thought we were talking about temperature. What has mass got
to do with it?

Nothing. He seems to believe that the kludge known as "relativistic
mass" somehow invalidates my argument.

This has been going on since April. He didn't understand then, and
doesn't understand now as evidenced by his latest little temper
tantrum. Every so often he has to spam this newsgroup with about 20
posts saying how stupid I am and whatever.

http://groups.google.com/group/sci.physics.relativity/msg/8ec41d88e75...

That is the argument for why temperature transforms as such.

I'm not so sure, anymore. Based on a quick search of arXiv, this
seems to be an unsettled area of physics, with some saying

T' = T * (1 - v^2/c^2)^(1/2),

some saying

T' = T / (1 - v^2/c^2)^(1/2),

and some saying

T' = T.

See arXiv:gr-qc/9505045, arXiv:physics/0506214, and
arXiv:physics/9610016, for instance.

gr-qc/9505045 : Assumes quantum field theory. Gives T' = gamma^3 * T.
Cute...

physics/0506214 : Asks the important question: "What should be kept
invariant under Lorentz transformation?!".

Different authors assume different things. For example, I follow
Tolman and assume that /entropy/ is invariant. ***, it makes sense to
me. Why would you have different amounts of available particle states
just because you are in a different reference frame? Other authors
assume that temperature should remain invariant with the same type of
argument, then they discover that entropy transforms instead.

I agree that no consensus will be reached until there is some actual
experimental evidence in one direction or the other.

physics/9610016 : Assumes quantum field theory...and finds that T' = T
for a blackbody thermal spectrum. This was actually touched upon in gr-
qc/9505045. I'd care more if I knew enough quantum field theory to
understand exactly what was being done.

This may sound stupid, but the result you obtain depends very strongly
on what you assume. QFT is not classical thermodynamics + SR, and the
results of both heavily depend on your starting assumptions. All sorts
of assumptions have a well-motivated physical basis, and none of them
can be preferred until we have a little more understanding.


That something so seemingly basic is still unsettled after 100
years suggests to me that there is something wrong with the
question. I suspect, but cannot prove, that T does not transform
at all (which is different from saying T' = T). I further
suspect that this is because a system in equilibrium in one frame
is not in equilibrium in another.

It doesn't surprise me at all to see that this is a somewhat
complicated subject which does not have a clear consensus.

However, the whole discussion wasn't started on some well-meant
academic debate. Back in APRIL [almost 4 fucking months ago] guskz
made one of his latest spew threads in which he misunderstands yet
another topic in physics. I was stupid, and tried to educate.

http://groups.google.com/group/sci.physics.relativity/browse_frm/thread/dd959050d1560e4

Eventually I end up pointing out that temperature is not a tensor
because tensors are invariant and temperature transforms under a
Lorentz transformation. Four months later, he is still pitching a
shitfit. My original point stands - temperature is not a tensor.

[...]


.