|
|
Posted by Neredbojias on 04/21/07 07:57
On Wed, 18 Apr 2007 20:31:07 GMT Bernhard Sturm scribed:
> Neredbojias wrote:
>> On Wed, 18 Apr 2007 08:24:22 GMT Bernhard Sturm scribed:
>>
>>> Neredbojias wrote:
>>>> On Tue, 17 Apr 2007 10:37:40 GMT Bernhard Sturm scribed:
>>>>
>>>>> Just a question: how big is the minimal TCP/IP packet-size? Do you
>>>>> really think it makes a difference if you transport 28 bytes
>>>>> instead of 128 bytes? With the former you will just have more
>>>>> overhead, that's all. IMHO This is purely an academic issue and
>>>>> not of real-world importance.
>>>> Years ago I played-around with packet-size at the larger end of the
>>>> scale and it made quite a difference on 56k.
>>>>
>>> so it makes a difference if I am going to send 28bytes in 4kb
>>> packets instead of 128bytes in 4kb packets?
>>> how is that?
>>
>> Doubtful, and even putting words in my mouth won't make it so.
>>
>> Now before you accuse me of obfuscation, re-read the first 2
>> sentences of your above post and see if they make you dizzy this
>> time.
>>
>
> sorry.. I am confused. What did I put in your mouth? I was just asking
> a question as you seemed to have experimented with packet-size. No
> intention of accusing you of anything. I thought you could share those
> results with us.
Well, I'm confused now, too, but it isn't that important so let's not worry
about it. How 'bout this: varying the size of packets makes a difference
but having less than full content in a packet makes little or none...
--
Neredbojias
He who laughs last sounds like an idiot.
Navigation:
[Reply to this message]
|