New Killernic K1

[dangel]

not to be blunt, but nobody serious about scientific computing uses ethernet - let alone tcp/ip. or windows. that's pretty much shortened the list.

...we do. So that's one sweeping generalisation nuked. Sure we use a canbus for the electronics but it's all windows monitored/controlled and we have plenty of TCP/IP based apps too..

[Nox]

ok, so 1 person Theres an exception to every rule, whats the chances of him being here too!!

/grin

Nox

[dangel]

ok, so 1 person Theres an exception to every rule, whats the chances of him being here too!!

/grin

Nox

Lol Well 'one' global multinational company anyway.. TBH all our competitors are Windows-based too. This is a stupid bit of hardware nonetheless..

[Nox]

and thats exactly what i'm hoping Hexus will prove either way. This seems to be the logical response, from anyone who knows anything about computers, theoretically it should help, but not by much.

but...

I'm waiting for Hexus to review, or for somewhere in the UK to sell them - until I actually own one or read a review here, i'll keep an open mind, especially when i've not seen any recent review say anything bad about them, bar the price!

Nox

[aidanjt]

I can't say it's bad hardware, just pointless, it wont make a real difference to gaming, nothing like the company claims.
The Intel PRO/1000 CT NIC performs TCP/UDP checksum offloading to the card, while it doesn't execute an entire TCP/IP stack onboard, it still helps during high load, and the card only costs 20-30 quid, and works on every operating system worth mentioning.

[directhex]

...we do. So that's one sweeping generalisation nuked. Sure we use a canbus for the electronics but it's all windows monitored/controlled and we have plenty of TCP/IP based apps too..

think we're using our own definitions of scientific computing

[aidanjt]

think we're using our own definitions of scientific computing

I was thinking of scientific instrumentation, same?

[dangel]

think we're using our own definitions of scientific computing

Well you stick to the one that suits you then.. What do you do for a living out of interest?

I was thinking of scientific instrumentation, same?

Me too, although we do a lot of algorithmic analysis etc too..

[Moby-Dick]

I know who he works for , and I think he does have a fairly good defenintion of scientific computing.

Currently we specialise in parallel, scientific High performance computing though we do allow serial applications onto our systems.

the above quote is from their website

[MadduckUK]

they may be all "scientific computing" but there websites a bit cack

[aidanjt]

Scientists tend to not care about their websites

[dangel]

I know who he works for , and I think he does have a fairly good defenintion of scientific computing.

Apparently not, if he's excluding Windows as a platform for it!

[Moby-Dick]

Its also Hexxehs he runs large scale parallel computing clusters which isn't really a common windows app

from what I've come accross, many parallel computing environments would use something like Myrinet low latency links between the nodes. ( have a search for Rhys's cluster project if you've not come accross it )

[dangel]

You're missing the point (and oddly I'm replying to you not him(?)) frankly - the field extends somewhat beyond the realms of server farms in the land of academia. Never mind..

[directhex]

You're missing the point (and oddly I'm replying to you not him(?)) frankly - the field extends somewhat beyond the realms of server farms in the land of academia. Never mind..

use of computers in science extends further than either of us, and certainly includes use of things like windows and bog standard ethernet

but i'd say there was something of a difference in using computers to perform science, and computers as a tool during science. wikipedia (yeah, i know, terrific place to cite) seems to agree with that

Scientific computing (or computational science) is the field of study concerned with constructing mathematical models and numerical solution techniques and using computers to analyze and solve scientific and engineering problems. In practical use, it is typically the application of computer simulation and other forms of computation to problems in various scientific disciplines.

instrumentation is doubtless an integral part of some fields, and simulation is an integral part of others. try telling the met office their sx-8's don't count for anything because they're not running windows - equally, try telling them they're not allowed to run antique win98 boxes out in their weather stations because they lack infiniband

at any rate, this is a silly discussion. it's fairly obvious, not only from context, what i was getting at

[dangel]

use of computers in science extends further than either of us, and certainly includes use of things like windows and bog standard ethernet

Agreed - but i'm talking about scientific computing - I thought you were too..

but i'd say there was something of a difference in using computers to perform science, and computers as a tool during science. wikipedia (yeah, i know, terrific place to cite) seems to agree with that

Let's just ignore you reached for wikipedia (although perhaps you missed http://en.wikipedia.org/wiki/List_of...lysis_software) .. We actually do both (by those definitions) and in Windows to boot.

at any rate, this is a silly discussion. it's fairly obvious, not only from context, what i was getting at

Perhaps, perhaps not - your original (quoted by myself) point was invalid in my eyes, and (as I said) was a sweeping generalisation at best, ignorance at worst..