By: juanrga (nospam.delete@this.juanrga.com), August 12, 2014 1:38 am
Room: Moderated Discussions
Michael S (already5chosen.delete@this.yahoo.com) on August 12, 2014 1:20 am wrote:
> juanrga (nospam.delete@this.juanrga.com) on August 11, 2014 7:28 pm wrote:
> > Michael S (already5chosen.delete@this.yahoo.com) on August 10, 2014 6:01 am wrote:
> > > juanrga (nospam.delete@this.juanrga.com) on August 10, 2014 5:34 am wrote:
> > > > Evidently the x86 tax is more noticeable on small phone-like cores but the tax doesn't magically
> > > > vanish for big cores (only reduces the amount by a factor of about 2x or 3x). This is the reason
> > > > why a 90W ARM SoC is able to offer 80--90% of the performance of a Haswell 140W Xeon.
> > >
> > > How it could be true when there exists neither 90W ARM SoC nor 140W Haswell Xeon?
> > > http://ark.intel.com/products/codename/42174/Haswell#@Server
> > >
> >
> > 90W ARM SoCs were presented recently
>
> Do you know the difference between paper and silicon? Or between estimates based on
> preproduction silicon/preproduction boards and measurements on production systems?
The projects are real: some are in design phase, others in testing phase, others in production phase, others are already shipping...
I find it very funny that your argument was not used when someone mentioned Skylake Xeons in his post.
> > and there are several E5 Haswell Xeons rated at 140W.
>
> E5 - unlikely. 130W - yes, but not 140W.
> There exist one odd 150W IvyB Xeon-E5 part. Not sure for what reason, but most
> likely it's made for super-expensive workstations rather than for servers.
>
> 140W E7 is likely, but Haswell-based E7 is still half a year off.
>
> >
> > There are also several 135W and 145W Haswell Xeons.
>
> Click on link. There are none.
> Haswell-based Xeon-E5 is likely to be launched in the 2nd half of September. Right now
> it is not available for public benchmarking, only for internal one. So, there are no
> performance figures to be taken as 100% in order to derive what constitutes 80-90%.
Xeon E5-1620 v3 --- 140W --- CM8064401973600/BX80644E51620V3
Xeon E5-1630 v3 --- 140W --- CM8064401614501
Xeon E5-1650 v3 --- 140W --- CM8064401548111/BX80644E51650V3
Xeon E5-1660 v3 --- 140W --- CM8064401909200
...
Xeon E5-2690 v3 --- 135W --- CM8064401439416/BX80644E52690V3
...
Xeon E5-2699 v3 --- 145W --- CM8064401739300
Only because you cannot purchase something in a store now doesn't mean that it doesn't exist.
> juanrga (nospam.delete@this.juanrga.com) on August 11, 2014 7:28 pm wrote:
> > Michael S (already5chosen.delete@this.yahoo.com) on August 10, 2014 6:01 am wrote:
> > > juanrga (nospam.delete@this.juanrga.com) on August 10, 2014 5:34 am wrote:
> > > > Evidently the x86 tax is more noticeable on small phone-like cores but the tax doesn't magically
> > > > vanish for big cores (only reduces the amount by a factor of about 2x or 3x). This is the reason
> > > > why a 90W ARM SoC is able to offer 80--90% of the performance of a Haswell 140W Xeon.
> > >
> > > How it could be true when there exists neither 90W ARM SoC nor 140W Haswell Xeon?
> > > http://ark.intel.com/products/codename/42174/Haswell#@Server
> > >
> >
> > 90W ARM SoCs were presented recently
>
> Do you know the difference between paper and silicon? Or between estimates based on
> preproduction silicon/preproduction boards and measurements on production systems?
The projects are real: some are in design phase, others in testing phase, others in production phase, others are already shipping...
I find it very funny that your argument was not used when someone mentioned Skylake Xeons in his post.
> > and there are several E5 Haswell Xeons rated at 140W.
>
> E5 - unlikely. 130W - yes, but not 140W.
> There exist one odd 150W IvyB Xeon-E5 part. Not sure for what reason, but most
> likely it's made for super-expensive workstations rather than for servers.
>
> 140W E7 is likely, but Haswell-based E7 is still half a year off.
>
> >
> > There are also several 135W and 145W Haswell Xeons.
>
> Click on link. There are none.
> Haswell-based Xeon-E5 is likely to be launched in the 2nd half of September. Right now
> it is not available for public benchmarking, only for internal one. So, there are no
> performance figures to be taken as 100% in order to derive what constitutes 80-90%.
Xeon E5-1620 v3 --- 140W --- CM8064401973600/BX80644E51620V3
Xeon E5-1630 v3 --- 140W --- CM8064401614501
Xeon E5-1650 v3 --- 140W --- CM8064401548111/BX80644E51650V3
Xeon E5-1660 v3 --- 140W --- CM8064401909200
...
Xeon E5-2690 v3 --- 135W --- CM8064401439416/BX80644E52690V3
...
Xeon E5-2699 v3 --- 145W --- CM8064401739300
Only because you cannot purchase something in a store now doesn't mean that it doesn't exist.