By: anon (anon.delete@this.anon.com), June 4, 2013 9:43 pm
Room: Moderated Discussions
Exophase (exophase.delete@this.gmail.com) on June 4, 2013 10:36 am wrote:
> anon (anon.delete@this.anon.com) on June 4, 2013 4:03 am wrote:
> > Why would you think that? Intel has not proven to be particularly honest when they have not had
> > the advantage -- which up to now has been "most of the time", when trying to compete with ARMs.
> >
>
> Intel's tests comparing Silvermont and Saltwell are more likely to be accurate than tests between
> an x86 and ARM processor. They don't have a strong incentive to make their current product look
> bad when it'll be selling for another six months.
It's not making the current product look bad. It's making future product look good.
Let's face it, anybody who is choosing between ARM and Atom in their next product already knows how current products stack up. They would have extensively tested them in their own labs. Some powerpoint slide is not going to change anything there.
So given Atom's "success" against ARM so far, they have an extremely strong incentive to present their next product in the best possible light over their current product.
> So I doubt they deliberately sabotaged it by
> using inferior binaries than they could have, using poorly representative test setups, etc.
I think it would be quite easy and quite possible to shine a light on the best aspects of the new CPU.
Let's just take a binary compiled for generic core2/opteron era cores, and use that for both cores. That's not poorly representative, because such situations would be common in the field (particularly in microserver environment). It is probably a better real world test than using a highly tuned in-order Atom specific binary.
However you can be sure that when comparing current generation Atom with ARM CPUs, they will be using their highly tuned specific binaries.
I'm not saying they did that in particular, but it is an example of how you can paint a misleading picture even while non of the individual strokes can be claimed as an outright lie.
That previous comparison of the old ARM core that Wilco showed is much more blatant dishonesty, so we've already established how far they're willing to go.
> anon (anon.delete@this.anon.com) on June 4, 2013 4:03 am wrote:
> > Why would you think that? Intel has not proven to be particularly honest when they have not had
> > the advantage -- which up to now has been "most of the time", when trying to compete with ARMs.
> >
>
> Intel's tests comparing Silvermont and Saltwell are more likely to be accurate than tests between
> an x86 and ARM processor. They don't have a strong incentive to make their current product look
> bad when it'll be selling for another six months.
It's not making the current product look bad. It's making future product look good.
Let's face it, anybody who is choosing between ARM and Atom in their next product already knows how current products stack up. They would have extensively tested them in their own labs. Some powerpoint slide is not going to change anything there.
So given Atom's "success" against ARM so far, they have an extremely strong incentive to present their next product in the best possible light over their current product.
> So I doubt they deliberately sabotaged it by
> using inferior binaries than they could have, using poorly representative test setups, etc.
I think it would be quite easy and quite possible to shine a light on the best aspects of the new CPU.
Let's just take a binary compiled for generic core2/opteron era cores, and use that for both cores. That's not poorly representative, because such situations would be common in the field (particularly in microserver environment). It is probably a better real world test than using a highly tuned in-order Atom specific binary.
However you can be sure that when comparing current generation Atom with ARM CPUs, they will be using their highly tuned specific binaries.
I'm not saying they did that in particular, but it is an example of how you can paint a misleading picture even while non of the individual strokes can be claimed as an outright lie.
That previous comparison of the old ARM core that Wilco showed is much more blatant dishonesty, so we've already established how far they're willing to go.