n-dimensional raytracer
 

Hi,
this is an example program :

https://github.com/toki78/RayMultiDim.git
https://github.com/toki78/CPURootBeer.git

Needs at least an 8-core cpu.

Best Regards
Thorsten

Re: n-dimensional raytracer
 

Thorsten Kiefer wrote:
> this is an example program :

For maximum clarity, Usenet posts should have all relevant information in
the body of the post proper.

> https://github.com/******/RayMultiDim.git
> https://github.com/******/CPURootBeer.git
>
> Needs at least an 8-core cpu [sic].

Well, I'm sure you are very proud.

What is this, is it even in Java, and why should we care?

I am rather interested in how you determined that this program
(it is a program, right?) needs an eight-core CPU.

Care to share?

But don't share here if the program isn't written in Java.

--
Lew

Re: n-dimensional raytracer
 

On Tue, 1 Jan 2013 10:21:40 -0800 (PST), Lew wrote:

> Thorsten Kiefer wrote:
>> this is an example program :
> For maximum clarity, Usenet posts should have all relevant information in
> the body of the post proper.
>> https://github.com/******/RayMultiDim.git
>> https://github.com/******/CPURootBeer.git

>> Needs at least an 8-core cpu [sic].
> Well, I'm sure you are very proud.

> What is this, is it even in Java, and why should we care?

> I am rather interested in how you determined that this program
> (it is a program, right?) needs an eight-core CPU.

> Care to share?

> But don't share here if the program isn't written in Java.

It's Java, albeit pretty horrible Java - Exceptions are quietly swallowed,
everything is public, no comments, and the parts of the program I looked at
look like one giant race condition, where bugs are treated with the good
old "sleep and retry in a moment" method.

I could also find no indication that this program would need a special kind
of CPU, other than it spawning 16 ThreadS. My best guess is that this is a
port of a C program by someone who doesn't understand Java very well yet,
and probably never programmed for a living.

Liebe Gruesse,
Joerg

--
Ich lese meine Emails nicht, replies to Email bleiben also leider
ungelesen.

Re: n-dimensional raytracer
 

Joerg Meier wrote:
> Lew wrote:
>> Thorsten Kiefer wrote:
>>> this is an example program :
>> For maximum clarity, Usenet posts should have all relevant information in
>> the body of the post proper.
>>> https://github.com/******/RayMultiDim.git
>>> https://github.com/******/CPURootBeer.git
>
>>> Needs at least an 8-core cpu [sic].
>
>> Well, I'm sure you are very proud.
>
>> What is this, is it even in Java, and why should we care?
>
>> I am rather interested in how you determined that this program
>> (it is a program, right?) needs an eight-core CPU.
>
>> Care to share?
>
>> But don't share here if the program isn't written in Java.
>
> It's Java, albeit pretty horrible Java - Exceptions are quietly swallowed,

Good news. Then it's on topic.

> everything is public, no comments, and the parts of the program I looked at
> look like one giant race condition, where bugs are treated with the good
> old "sleep and retry in a moment" method.
>
> I could also find no indication that this program would need a special kind
> of CPU, other than it spawning 16 ThreadS. My best guess is that this is a
> port of a C program by someone who doesn't understand Java very well yet,
> and probably never programmed for a living.

So let's punt back to the OP: What would you like from us with regard to
your post?

Are you looking for comments? Advice?

--
Lew

Re: n-dimensional raytracer
 

Am 01.01.2013 23:17, schrieb Lew:
> Joerg Meier wrote:
>> Lew wrote:
>>> Thorsten Kiefer wrote:
>>>> this is an example program :
>>> For maximum clarity, Usenet posts should have all relevant information in
>>> the body of the post proper.
>>>> https://github.com/******/RayMultiDim.git
>>>> https://github.com/******/CPURootBeer.git
>>
>>>> Needs at least an 8-core cpu [sic].
>>
>>> Well, I'm sure you are very proud.
>>
>>> What is this, is it even in Java, and why should we care?
>>
>>> I am rather interested in how you determined that this program
>>> (it is a program, right?) needs an eight-core CPU.
>>
>>> Care to share?
>>
>>> But don't share here if the program isn't written in Java.
>>
>> It's Java, albeit pretty horrible Java - Exceptions are quietly swallowed,
>
> Good news. Then it's on topic.
>
>> everything is public, no comments, and the parts of the program I looked at
>> look like one giant race condition, where bugs are treated with the good
>> old "sleep and retry in a moment" method.
>>
>> I could also find no indication that this program would need a special kind
>> of CPU, other than it spawning 16 ThreadS. My best guess is that this is a
>> port of a C program by someone who doesn't understand Java very well yet,
>> and probably never programmed for a living.
>
> So let's punt back to the OP: What would you like from us with regard to
> your post?
>
> Are you looking for comments? Advice?
>

Yes, I would like some comments :)
The reason, why I said it needs an 8-core, is such that it renders in
realtime.

Regards
Thorsten

Re: n-dimensional raytracer
 

Thorsten Kiefer wrote:
> schrieb Lew:
>> Joerg Meier wrote:
>>> Lew wrote:
>>>> Thorsten Kiefer wrote:
>>>>> this is an example program :
>
>>>> For maximum clarity, Usenet posts should have all relevant information in
>>>> the body of the post proper.
>
>>>>> https://github.com/******/RayMultiDim.git
>>>>> https://github.com/******/CPURootBeer.git
>
>>>>> Needs at least an 8-core cpu [sic].
>
[snip]

>>>> I am rather interested in how you determined that this program
> >>> (it is a program, right?) needs an eight-core CPU.
>
> [snip]
>
>>> It's Java, albeit pretty horrible Java - Exceptions are quietly swallowed,
>
>> Good news. Then it's on topic.
>
>>> everything is public, no comments, and the parts of the program I looked at
>>> look like one giant race condition, where bugs are treated with the good
>>> old "sleep and retry in a moment" method.
>
>>> I could also find no indication that this program would need a special kind
>>> of CPU, other than it spawning 16 ThreadS. My best guess is that this is a
>>> port of a C program by someone who doesn't understand Java very well yet,
>>> and probably never programmed for a living.
>
>> So let's punt back to the OP: What would you like from us with regard to
>> your post?
>> Are you looking for comments? Advice?
>
> Yes, I would like some comments :)

Be careful what you wish for.

> The reason, why I said it needs an 8-core, is such that it renders in
> realtime.

How did you relate that to number of cores?

Your code needs an awful lot of work. You've got source that does nothing, Swing GUIs
started off the EDT, non-conventional variable names and indentation, and the issues Joerg
mentioned.

Check out the Java online tutorials at oracle.com and snoop around there for the Java
coding conventions.

If you make a request to the newsgroup, you have to actually state what the request is.

Thank you for responding to our questions.

--
Lew

Re: n-dimensional raytracer
 

On 1/2/2013 5:04 AM, Thorsten Kiefer wrote:
> Am 01.01.2013 23:17, schrieb Lew:
>> Joerg Meier wrote:
>>> Lew wrote:
>>>> Thorsten Kiefer wrote:
>>>>> Needs at least an 8-core cpu [sic].
>>>> I am rather interested in how you determined that this program
>>>> (it is a program, right?) needs an eight-core CPU.
> The reason, why I said it needs an 8-core, is such that it renders in
> realtime.

So it will render in realtime with 8 core no matter
the speed of those core or what ISA they are?

How do you achieve that in Java? Test on number
of cores?

Arne

Re: n-dimensional raytracer
 

On Tuesday, January 1, 2013 10:09:53 AM UTC-6, Thorsten Kiefer wrote:
> Hi,
>
> this is an example program :
>
>
>
> https://github.com/toki78/RayMultiDim.git
>
> https://github.com/toki78/CPURootBeer.git
>
>
>
> Needs at least an 8-core cpu.
>
>
>
> Best Regards
>
> Thorsten

If it is n-dimensional, what values of n make sense other than 3?

Re: n-dimensional raytracer
 

bob smith <bob@coolfone.comze.com> writes:
>If it is n-dimensional, what values of n make sense other than 3?

»We demonstrate our method on scenes with a 3 to 5
dimensional space«

hachisuka08multidimensional.pdf

Re: n-dimensional raytracer
 

Stefan Ram wrote:
> bob smith writes:
>> If it is n-dimensional, what values of n make sense other than 3?
>
> »We demonstrate our method on scenes with a 3 to 5
> dimensional space«
>
> hachisuka08multidimensional.pdf

We should google for this, eh?

--
Lew