Turn on suggestions

Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

- AMD Community
- Communities
- Developers
- Devgurus Archives
- Archives Discussions
- Is HD 5870 genuine 64 bit?

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Mute
- Printer Friendly Page

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content

03-17-2012
10:37 PM

Is HD 5870 genuine 64 bit?

Does the ALU of HD 5870 support native 64 bit calculations? Does the following instructions use the same cycles?

long int x, y, z;

z = x + y;

int a, b, c;

c = a + b;

Thank you in advance!

Solved! Go to Solution.

1 Solution

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content

03-18-2012
04:41 PM

Although add/subtract run at the full rate, multiply isn't so lucky.

The programming guide, section 4.13.1 has a bit about data type performance - note that these things aren't even really '32 bit' integer devices either, and a 32 bit multiply has 1/5 the rate of a 24 bit or float multiply.

Presumably all long ops are implemented using 32 bit ones.

11 Replies

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content

03-18-2012
12:48 PM

Does the ALU of HD 5870 support native 64 bit calculations?

It is not native 64 bit, but can do 64 bit calculations.

Does the following instructions use the same cycles?

AFAIK Double Precision Operations need 5 times more cycles than Single Precision Operations.

--

Srdja

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content

03-18-2012
01:14 PM

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content

03-22-2012
07:57 PM

Thank you settle! Very helpful answer!

I have to use a simple 64-bit calculation. I think an on-chip emulation is more efficient than a home-made emulation in C-language

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content

03-22-2012
07:53 PM

thank you smatovic for your helpful answer!

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content

03-18-2012
04:41 PM

Although add/subtract run at the full rate, multiply isn't so lucky.

The programming guide, section 4.13.1 has a bit about data type performance - note that these things aren't even really '32 bit' integer devices either, and a 32 bit multiply has 1/5 the rate of a 24 bit or float multiply.

Presumably all long ops are implemented using 32 bit ones.

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content

03-19-2012
07:13 PM

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content

03-22-2012
08:02 PM

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content

03-22-2012
08:00 PM

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content

03-22-2012
09:21 PM

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content

03-22-2012
09:25 PM

hmm, so e-mail replying doesn't seem to work at all. blah.

I just stated that the 24-bit multiply uses the single-floating point mantissa multipliers, which are only 24-bit. So it makes sense when the cards are trying to pack as much float performance in as possible. It's still a useful size for many integer algorithms, and particularly address calculations.

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content

03-23-2012
05:14 PM

Thank you notzed. You just brought 24-bit numbers to my attention