# glCompressedTexSubImage2D()

**URL:** https://forums.imgtec.com/t/glcompressedtexsubimage2d/168
**Category:** PowerVR Insider
**Created:** [October 17, 2008, 6:50am UTC](https://forums.imgtec.com/t/glcompressedtexsubimage2d/168 "2008-10-17T06:50:19Z")
**Posts on this page:** 6
**Page:** 1

<div class="post-metadata">

### Author: ![iseno](https://avatars.discourse-cdn.com/v4/letter/i/c37758/32.png) [@iseno](https://forums.imgtec.com/u/iseno)
#### Post date: [October 17, 2008, 6:50am UTC](https://forums.imgtec.com/t/glcompressedtexsubimage2d/168/1 "2008-10-17T06:50:19Z")

</div>

Hi,
  

&nbsp;
  

I support developping Embedded Graphics System.
  

I use OpenGL ES 1.1 on&nbsp;Freescale i.MX31 (ARM11+MBX 533MHz).
  

&nbsp;
  

I&nbsp;want to know if MBX's OpenGLES 1.1 supports " glCompressedTexSubImage2D()".
  

If it is true, I want the detail about this function.\<?: prefix = o ns = "urn:schemas-microsoft-com:office:office" /\>

  

Because I cannot find any information on public document as follows.

  

[http://www.khronos.org/opengles/documentation/opengles1\_0/html/glCompressedTexSubImage2D.html](http://www.khronos.org/opengles/documentation/opengles1_0/html/glCompressedTexSubImage2D.html)

  

  

&nbsp;

  

* * *
\>

  

&nbsp;format

  

&nbsp;Specifies the format of the pixel data. Currently, there is no supported format.

  

&nbsp;

  

Can anyone help me?

---

<div class="post-metadata">

### Author: ![CarlosSarria](https://sea1.discourse-cdn.com/flex015/user_avatar/forums.imgtec.com/carlossarria/32/263_2.png) [@CarlosSarria](https://forums.imgtec.com/u/CarlosSarria)
#### Post date: [October 17, 2008, 4:15pm UTC](https://forums.imgtec.com/t/glcompressedtexsubimage2d/168/2 "2008-10-17T16:15:44Z")

</div>

Hello,

  

No, MBX does not supports glCompressedTexSubImage2D(). Texture compression is a very expensive operation that cannot be done on the fly. Besides, It is not possible just to upload part of a PVRTC texture as all adjacent blocks contribute to the colour.

  

Regards.

  

Carlos.

---

<div class="post-metadata">

### Author: ![iseno](https://avatars.discourse-cdn.com/v4/letter/i/c37758/32.png) [@iseno](https://forums.imgtec.com/u/iseno)
#### Post date: [October 22, 2008, 4:12am UTC](https://forums.imgtec.com/t/glcompressedtexsubimage2d/168/3 "2008-10-22T04:12:01Z")

</div>

Hi Carlos,\<?: prefix = o ns = "urn:schemas-microsoft-com:office:office" /\>

  

&nbsp;

  

I am sorry my late response.

  

Thanks for your information. This is very important.

  

&nbsp;

  

I've never heard that MBX is NOT support glCompressedTexSubImage2D() from the vendor in \<?: prefix = st1 ns = "urn:schemas-microsoft-com:office:smarttags" /\>Japan.

  

&nbsp;

  

&nbsp;

  

&nbsp;Akihiro Iseno

  

&nbsp;TOMEN Electronics Corp. \<http://www.tomen-ele.co.jp\>

  

&nbsp;&nbsp;&nbsp;&nbsp; Engineering Dept.

  

&nbsp;&nbsp;&nbsp;&nbsp; Amusement SBU, Sales Division II

---

<div class="post-metadata">

### Author: ![iseno](https://avatars.discourse-cdn.com/v4/letter/i/c37758/32.png) [@iseno](https://forums.imgtec.com/u/iseno)
#### Post date: [October 22, 2008, 6:55am UTC](https://forums.imgtec.com/t/glcompressedtexsubimage2d/168/4 "2008-10-22T06:55:32Z")

</div>

Hi Carlos,\<?: prefix = o ns = "urn:schemas-microsoft-com:office:office" /\>

  

&nbsp;

  

You said that "Texture compression is a very expensive operation that cannot be done on the fly."

  

&nbsp;

  

Does MBX support glCompressedTexImage2D() operation?

  

Which operation is speedy, glCompressedTexImage2D() or glTexImage2D() on MBX?

  

&nbsp;

  

&nbsp;&nbsp;&nbsp;Akihiro Iseno

---

<div class="post-metadata">

### Author: ![CarlosSarria](https://sea1.discourse-cdn.com/flex015/user_avatar/forums.imgtec.com/carlossarria/32/263_2.png) [@CarlosSarria](https://forums.imgtec.com/u/CarlosSarria)
#### Post date: [October 22, 2008, 3:23pm UTC](https://forums.imgtec.com/t/glcompressedtexsubimage2d/168/5 "2008-10-22T15:23:29Z")

</div>

Dear Iseno-san,  
Texture compression is done offline by our PVRTC Utility. The API uses glCompressedTexImage2D just to upload the pre-compressed data in memory, so it is as fast as glTexImage2D.  
The user cannot not modify this data on any way because it is expensive to re-compress so glCompressedTexSubImage2D is not supported. If this kind of functionality is required please use a 16-bit texture format.

  

Best regards.

  

Carlos.

---

<div class="post-metadata">

### Author: ![iseno](https://avatars.discourse-cdn.com/v4/letter/i/c37758/32.png) [@iseno](https://forums.imgtec.com/u/iseno)
#### Post date: [October 23, 2008, 1:15am UTC](https://forums.imgtec.com/t/glcompressedtexsubimage2d/168/6 "2008-10-23T01:15:19Z")

</div>

Dear Carlos-san,\<?: prefix = o ns = "urn:schemas-microsoft-com:office:office" /\>

  

&nbsp;

  

Thanks for good information.

  

I will try to do it. :-)

  

&nbsp;

  

Yours sincerely,

  

&nbsp;

  

Akihiro Iseno

  

&nbsp;TOMEN Electronics Corp. \<http://www.tomen-ele.co.jp\>

  

&nbsp;&nbsp;&nbsp;&nbsp; Engineering Dept.

  

&nbsp;&nbsp;&nbsp;&nbsp; Amusement SBU, Sales Division II