[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Public WebGL] GL_MAX_TEXTURE_SIZE power of two ?



I don't think 'k' is usually lod, and in particular 'k' is not lod in that passage, as 'lod' is lod. 'k' does generally appear as an integer in the spec, but it's usually unambiguously specified as such. As written, the passage in question works fine as a continuous formula, restricted in practice by the inability to specify fractional texture levels.

I personally find the argument that MAX_TEXTURE_SIZE is coerced to POT by this passage weak. If we want to specify it as such, we should just do so explicitly. The spec should be clear, and it's always a little embarrassing when something as simple as this results in such subtly-positioned arguments.

I think it is much more useful to allow NPOT MAX_TEXTURE_SIZE, since we require NPOT texture support anyways. It would be a useful hint if an implementation can support 7k^2 textures, but not quite 8k^2 textures, rather than having to drop back all the way to 4k^2. Without going into details on this forum, there are a number of implementations which would (appear to*) suffer from forcing MAX_TEXTURE_SIZE to POT.

* I have a post I need to write about this tomorrow. In short, I can't find the language in the spec that says TexImage2D(2*MAX_TEXTURE_SIZE,1) should trigger INVALID_VALUE. Props if anyone can find it before then.

-Jeff



----- Original Message -----
From: "Mark Callow" <callow.mark@artspark.co.jp>
To: "Kenneth Russell" <kbr@google.com>
Cc: "Guillaume Abadie" <guillaume.abadie@gmail.com>, "Brandon Jones" <bajones@google.com>, "public webgl" <public_webgl@khronos.org>
Sent: Monday, July 15, 2013 9:01:55 PM
Subject: Re: [Public WebGL] GL_MAX_TEXTURE_SIZE power of two ?

On 2013/07/16 5:01, Kenneth Russell wrote:
> Personally I'm not sure that a specification and conformance suite
> update is warranted for this very basic assumption. Please feel free
> to send pull requests to https://github.com/KhronosGroup/WebGL for
> review.
The OpenGL ES 2.0 spec. repeatedly refers to k as a level of detail. The
GL certainly does not have fractional levels of detail so it is pretty
darn clear that k is an integer and therefore MAX_{WIDTH,HEIGHT} are
powers of two.

I don't know where those 151 people with 16320 came from. There's
probably a bug somewhere.

I agree with Ken. The only change I think remotely warranted is to have
the CTS test that MAX_{WIDTH,HEIGHT} are powers of two.

Regards

    -Mark

-- 
注意:この電子メールには、株式会社エイチアイの機密情報が含まれている場合
が有ります。正式なメール受信者では無い場合はメール複製、 再配信または情
報の使用を固く禁じております。エラー、手違いでこのメールを受け取られまし
たら削除を行い配信者にご連絡をお願いいたし ます.

NOTE: This electronic mail message may contain confidential and
privileged information from HI Corporation. If you are not the intended
recipient, any disclosure, photocopying, distribution or use of the
contents of the received information is prohibited. If you have received
this e-mail in error, please notify the sender immediately and
permanently delete this message and all related copies.


-----------------------------------------------------------
You are currently subscribed to public_webgl@khronos.org.
To unsubscribe, send an email to majordomo@khronos.org with
the following command in the body of your email:
unsubscribe public_webgl
-----------------------------------------------------------