• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

AMD Zen 2 Offers a 13% IPC Gain over Zen+, 16% over Zen 1

No competition results in increased price, not lack of performance.

No competition often sees a lack of investment in innovation or progress artificially slowed in roll out to maximise profit which tends to result in smaller performance jumps. There is no hard and fast rule in that respect though.

EDIT: Not sure why I'm even explaining that we've decades of history showing various outcomes when companies do and don't have competition.
 
No competition often sees a lack of investment in innovation or progress artificially slowed in roll out to maximise profit which tends to result in smaller performance jumps. There is no hard and fast rule in that respect though.

EDIT: Not sure why I'm even explaining that we've decades of history showing various outcomes when companies do and don't have competition.

One example of this would be the electric car market, when you get companies like Tesla coming from no where and starting to eat into the market share of the incumbents, all of a sudden they care again and have a whole range coming out themselves to compete.
 
I wonder if a tax on cash piles above a certain amount would promote continued investment and development rather than just sitting pretty. The amount of cash that Apple has just doing nothing is obscene, for example.
 
Doesn't fit this context his second quote says they've done very little - which is what I was saying - without much competition over the last 10 years they've not particularly pushed themselves to wring out IPC gains so not really much of a metric for what good progress should look like. They might be trying now but that is irrelevant.
I think Intel management has all the time planned architectural improvement (Ice Lake) for 10nm node and failed to take into account how bad SNAFU 10nm is.
So lacking back up plan, they've just had to keep respinning Skylake every time they've been forced to admit that 10nm still doesn't work.
If they had been purposely holding back architectural improvement, now would have been good time to bring it.
 
I wonder if a tax on cash piles above a certain amount would promote continued investment and development rather than just sitting pretty. The amount of cash that Apple has just doing nothing is obscene, for example.
Would they just not distribute it via dividends then?

Though I am sure it would encourage them to invest more also.
 
I know - 7nm gives a huge amount of potential for improvements over "12nm".

But aren't IPC improvements pretty much entirely down to changes to the architecture / design - the specification of exactly what happens logic wise per clock? While the manufacturing process simply determines how much circuitry you can build, how much power it needs to run, how fast transistors can switch etc.
 
But aren't IPC improvements pretty much entirely down to changes to the architecture / design - the specification of exactly what happens logic wise per clock? While the manufacturing process simply determines how much circuitry you can build, how much power it needs to run, how fast transistors can switch etc.

Changes to the material process allow for changes to the architecture - as parts of the design can now potentially switch faster (potentially allows for more prediction misses inside a clock cycle, etc.), physically closer which reduces the effects of noise and other parasitics and potentially latency and so on that enables you to do things that before weren't possible, utilise caches in a way that wasn't possible before and/or increase the amount of cache local to certain areas of the CPU and so on - stuff like this can have a huge impact on what is possible with things like branch prediction increase the effectiveness of your CPU per clock cycle.
 
But aren't IPC improvements pretty much entirely down to changes to the architecture / design - the specification of exactly what happens logic wise per clock? While the manufacturing process simply determines how much circuitry you can build, how much power it needs to run, how fast transistors can switch etc.

Yes.
 

No - IPC comes from architecture changes, but you often need progress in the semiconductor technology to make evolving designs possible - some implementations to make the CPU work more efficiently simply aren't possible without being backed up by the material tech when it comes to the latency and cache potential and so on.
 
When is Zen2 due out ? I might wait if its not far off rather than buy the ridiculous 9900k. Although I think Zen2 will be just as much imo.
 
But the dies should be smaller, so more dies per wafer often offsets that cost.

For a comparative product yeah but 7nm is a pretty big jump in costs so it won't be offset as much as it has been sometimes in the past. Design costs increase about 3x over 14/16nm, initial wafer cost will be 2.5-3x though that will come down by 2021.
 
For a comparative product yeah but 7nm is a pretty big jump in costs so it won't be offset as much as it has been sometimes in the past. Design costs increase about 3x over 14/16nm, initial wafer cost will be 2.5-3x though that will come down by 2021.

Wafer density will also be about 2.5x - 3x as well though.
 
Wafer density will also be about 2.5x - 3x as well though.

Yeah but in the past it would have been more typically in the region of 1.3x not 2.5-3x cost increase. Though 14/16 finfet cost were quite a big jump as well so it won't suddenly ramp quite as dramatically from current prices as it could have done.
 
Back
Top Bottom