3 ms·
Funny, I think this graph is from a thread where I got roasted for arguing that differences in CPU implementations meant that Intel (or anyone else really) woul
by creato 6y ago
Funny, I think this graph is from a thread where I got roasted for arguing that differences in CPU implementations meant that Intel (or anyone else really) would need to be careful about shipping SIMD software on processors they don't test.
- Dylan16807 6y agoRightfully so. Because they were replacing the SIMD code with non-SIMD code that was similarly untested, with the side effect of crippling performance on their competitor's chips. That's a really bad kind of "being careful".
- creato 6y agoI'm not aware of instructions that have this implementation defined behavior risk in non-SIMD code, and you can be sure that Intel has tested this. If Microsoft had shipped a non-SIMD version of D3DXMatrixInverse, this bug likely could not have existed (or it would have been caught on the developer's machines). In this case (games), they probably made the right tradeoff despite the bug. But in general? I don't want to rehash the argument, I'm just really, really glad I'm not the one flipping that switch (or subject to the hordes of HNers who think I'm a monster for not flipping it).
- Const-me 6y agoCorrect, I've scavenged the comment, and the data+code, from an old thread here on HN.