They do still seem to be making advances in single-core performance, but whether it matters to most people is a different question. Most people aren’t using software that would benefit that much from these generation-to-generation performance improvements. It’s not going to be anywhere near as noticeable as when we went from 2 or 4 cores to 8, 16, 24, etc.
Single-thread is really hard, we’ve basically saturated our l1 working set size, adding more doesn’t help much. Trying to extend the vector length just makes physical design harder and that reduces clock speed. The predictors are pretty good, and Apple finally kicked everyone up the ass to increase OOO like they should have.
Also, software still kind of sucks. It’s better than it was, but we need to improve it, the bloat is just barely being handled by silicon gains.
Flash was the epochal change, maybe we have some new form of hybrid storage but that doesn’t seem likely right now, Apple might do it to cut costs while preserving performance, actually yeah I see them trying to have their cake and eat it too.
Otherwise I don’t know, we need a better way to deal with GPUs, there’s nothing else that can move the needle, except true heterogenous core clusters, but I haven’t been able to sell that to anyone so far, they all think it’s a great idea, that someone else should do.
Also, software still kind of sucks. It’s better than it was, but we need to improve it, the bloat is just barely being handled by silicon gains.
The incentives are all wrong for this, except in FOSS. It’s never going to be a priority for Microsoft because everyone is used to the (lack of) speed of Windows etc., and “now a bit faster!” isn’t a great marketing line. And it’s not in the interests of hardware companies that need to keep shifting new boxes if the software doesn’t keep bogging each generation down eventually. So we end up stuck with proprietary bloatware everywhere.
Let’s be fair, Ms was vastly outrunning Intel for a long time, it’s only slowed down recently, and now the problem isn’t single-thread bloat so much as it is an absolute lack of multicore scaling for almost all applications except some games, and even then windows fights as hard as it possibly can to stop you, like amd just proved yet again.
They do still seem to be making advances in single-core performance, but whether it matters to most people is a different question. Most people aren’t using software that would benefit that much from these generation-to-generation performance improvements. It’s not going to be anywhere near as noticeable as when we went from 2 or 4 cores to 8, 16, 24, etc.
Single-thread is really hard, we’ve basically saturated our l1 working set size, adding more doesn’t help much. Trying to extend the vector length just makes physical design harder and that reduces clock speed. The predictors are pretty good, and Apple finally kicked everyone up the ass to increase OOO like they should have.
Also, software still kind of sucks. It’s better than it was, but we need to improve it, the bloat is just barely being handled by silicon gains.
Flash was the epochal change, maybe we have some new form of hybrid storage but that doesn’t seem likely right now, Apple might do it to cut costs while preserving performance, actually yeah I see them trying to have their cake and eat it too.
Otherwise I don’t know, we need a better way to deal with GPUs, there’s nothing else that can move the needle, except true heterogenous core clusters, but I haven’t been able to sell that to anyone so far, they all think it’s a great idea, that someone else should do.
The incentives are all wrong for this, except in FOSS. It’s never going to be a priority for Microsoft because everyone is used to the (lack of) speed of Windows etc., and “now a bit faster!” isn’t a great marketing line. And it’s not in the interests of hardware companies that need to keep shifting new boxes if the software doesn’t keep bogging each generation down eventually. So we end up stuck with proprietary bloatware everywhere.
“what intel gives, microsoft takes away”
dates from the mid 90s, still relevant.
Let’s be fair, Ms was vastly outrunning Intel for a long time, it’s only slowed down recently, and now the problem isn’t single-thread bloat so much as it is an absolute lack of multicore scaling for almost all applications except some games, and even then windows fights as hard as it possibly can to stop you, like amd just proved yet again.