 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I added a 2.8Ghz 6 core i5 to the farm this week.
( http://www.buckosoft.com/bsac/meta/ )
$500 including a nice 23" HD monitor.
I expected it to be slighty faster than my 3.4 Ghz i7 (4 core/8 thread)
and 50% faster than my (3) 3.2Ghz 4 core i5s.
No.
It is 50% faster than my i7 and more than twice as fast as my other i5(s).
I'm not complaining ( :) ) but sheesh. Shouldn't intel call this an i6
or something? Clearly it is a newer/faster breed than the other boxes
that have an i5. It seems that intel is really underselling themselves
here.
--
dik
Rendered 920576 of 921600 pixels (99%)
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Am 21.08.2018 um 05:26 schrieb dick balaska:
> I'm not complaining ( :) ) but sheesh. Shouldn't intel call this an i6
> or something? Clearly it is a newer/faster breed than the other boxes
> that have an i5. It seems that intel is really underselling themselves
> here.
The digit behind the "i" has nothing to do with absolute computing
power, and instead denotes the marketing segment the CPU is designed
for: office/low-end (i3), mid-range (i5), high-end (i7) or
super-high-end (i9). As CPU power increases in general over time, so
does the CPU power in each marketing segment.
You /can/ be reasonably sure that e.g. a brand-spanking-new i7
- does more computations per second
- eats up more Watts per computation
- eats up more dollars per computations per second
than an equally brand-spanking-new i5, but that's about it.
Intel has moved away from trying to convey absolute computing power in
their CPU names; and rightly so, I think, because the performance of
modern CPUs greatly depends on the application to be executed: For some
applications brute integer computing power is key, for others it is
brute floating point computing power, for yet others it is vector
computing power, and for some it is branch prediction; for some the
speed of the cache is key, for others it is the size of the cache, for
yet others it is sequential access RAM performance, and for some it is
random access RAM performance; for some the speed of the individual
cores is key, for others it is the combined speed of the physical cores,
and some benefit from hyperthreading while others don't. There's so much
you can optimize in a modern CPU to get better performance for /some/
applications that the idea of naming CPUs according to their performance
is moot these days.
Not to mention that pure computing speed has long ceased to be
everything that matters in comparing CPUs. Fiscal efficiency - i.e. how
many computations per second you get per buck - has probably been a
factor as long as computers exist, and energy efficiency - i.e. how many
computations you get per Watt - is becoming ever more important.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Le 21/08/2018 à 05:26, dick balaska a écrit :
> I added a 2.8Ghz 6 core i5 to the farm this week.
> ( http://www.buckosoft.com/bsac/meta/ )
> $500 including a nice 23" HD monitor.
>
>
> I expected it to be slighty faster than my 3.4 Ghz i7 (4 core/8 thread)
> and 50% faster than my (3) 3.2Ghz 4 core i5s.
>
> No.
>
> It is 50% faster than my i7 and more than twice as fast as my other i5(s).
>
> I'm not complaining ( :) ) but sheesh. Shouldn't intel call this an i6
> or something? Clearly it is a newer/faster breed than the other boxes
> that have an i5. It seems that intel is really underselling themselves
> here.
>
Was this speed measured with povray benchmark ?
Same binary ?
What about the memory speed ? (and CL values)
Beware of the cache effect.
From an old i7 measure (on i7-980X), thread without core are between
20% and 25% of thread with core, when rendering with 3.7.
Not only the new generation is faster, it is also smaller and use less
power. Yet no one wants to break the 4GHz barriers.
(but on the other hand, the new generation has a backdoor on the
network, even in suspended state and it cannot be closed)
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Am 21.08.2018 um 13:08 schrieb Le Forgeron:
> From an old i7 measure (on i7-980X), thread without core are between 20%
> and 25% of thread with core, when rendering with 3.7.
I guess you mean, "thread with virtual core" (i.e. hyperthreading) and
"thread with physical core".
A literal "thread without core" (setting `+wt` higher than the number of
virtual cores) will typically give you 0% performance gain (or even a
performance loss) in POV-Ray.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 08/21/2018 07:08 AM, Le Forgeron wrote:
> Was this speed measured with povray benchmark ?
> Same binary ?
Same binary. Not the benchmark, building frames for my animation.
>
> What about the memory speed ? (and CL values)
> Beware of the cache effect.
I don't care about any of that. All I care is that POV-Ray finishes
twice as fast. :)
>
> From an old i7 measure (on i7-980X), thread without core are between 20%
> and 25% of thread with core, when rendering with 3.7.
Yes, I typically see around 20% increase in throughput with my i7 vs. an
i5. It *can* float to near 100% on occasion, depending on what I'm
looking at. One thing I always thought odd, my kitchen volcano scene,
where I'm looking at the floor (superellipsoid tiles with a box for
grout) the i7 is actually 5-10% slower than an i5.
(This floor: http://www.buckosoft.com/tteoac/video/frames/ttko0524.png )
>
> Not only the new generation is faster, it is also smaller and use less
> power. Yet no one wants to break the 4GHz barriers.
I've noticed a Ghz decrease in my last couple of boxes. From 3.4Ghz to
3.0Ghz and now 2.8Ghz. Certainly that will run "cooler".
The box is in the living room and is silent, as opposed to my i7 which
has dual fans that have a teeth-floating harmonic.
> (but on the other hand, the new generation has a backdoor on the
> network, even in suspended state and it cannot be closed)
um...
--
dik
Rendered 920576 of 921600 pixels (99%)
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 8/20/2018 11:26 PM, dick balaska wrote:
> I added a 2.8Ghz 6 core i5 to the farm this week.
> ( http://www.buckosoft.com/bsac/meta/ )
> $500 including a nice 23" HD monitor.
>
$500 for the whole computer, or just the CPU + monitor?
Mike
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 08/21/2018 11:03 PM, Mike Horvath wrote:
> On 8/20/2018 11:26 PM, dick balaska wrote:
>> I added a 2.8Ghz 6 core i5 to the farm this week.
>> ( http://www.buckosoft.com/bsac/meta/ )
>> $500 including a nice 23" HD monitor.
>>
>
> $500 for the whole computer, or just the CPU + monitor?
>
>
> Mike
Best Buy. $519. 12GB RAM 1TB hard drive.
--
dik
Rendered 920576 of 921600 pixels (99%)
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 8/22/2018 2:27 AM, dick balaska wrote:
> On 08/21/2018 11:03 PM, Mike Horvath wrote:
>> On 8/20/2018 11:26 PM, dick balaska wrote:
>>> I added a 2.8Ghz 6 core i5 to the farm this week.
>>> ( http://www.buckosoft.com/bsac/meta/ )
>>> $500 including a nice 23" HD monitor.
>>>
>>
>> $500 for the whole computer, or just the CPU + monitor?
>>
>>
>> Mike
>
> Best Buy. $519. 12GB RAM 1TB hard drive.
>
Do you have a link? There are some good deals at BB in a "salvage parts
from an old PC to help build a new one" sense. I.e. I have an SSD and
graphics card already that I could put into one of these machines.
Mike
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 08/22/2018 06:18 PM, Mike Horvath wrote:
> On 8/22/2018 2:27 AM, dick balaska wrote:
>> On 08/21/2018 11:03 PM, Mike Horvath wrote:
>>> On 8/20/2018 11:26 PM, dick balaska wrote:
>>>> I added a 2.8Ghz 6 core i5 to the farm this week.
>>>> ( http://www.buckosoft.com/bsac/meta/ )
>>>> $500 including a nice 23" HD monitor.
>>>>
>>>
>>> $500 for the whole computer, or just the CPU + monitor?
>>>
>>>
>>> Mike
>>
>> Best Buy. $519. 12GB RAM 1TB hard drive.
>>
>
> Do you have a link? There are some good deals at BB in a "salvage parts
> from an old PC to help build a new one" sense. I.e. I have an SSD and
> graphics card already that I could put into one of these machines.
>
>
> Mike
Pretty weird. This is the box I bought, the link is from my history.
But, I did *not* pay $800 with monitor. I would have gone for a lesser
machine.
Technically, the machine is for the 6 year old to play minecraft. :)
https://www.bestbuy.com/site/dell-inspiron-desktop-intel-core-i5-12gb-memory-1tb-hard-drive-black-with-silver-trim/6228201.p?skuId=6228201
--
dik
Rendered 920576 of 921600 pixels (99%)
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 08/23/2018 06:07 AM, dick balaska wrote:
>
>
https://www.bestbuy.com/site/dell-inspiron-desktop-intel-core-i5-12gb-memory-1tb-hard-drive-black-with-silver-trim/6228201.p?skuId=6228201
>
"Smart six-core, twelve-way processing performance"
No it doesn't.
$ cat /proc/cpuinfo
vendor_id : GenuineIntel
model name : Intel(R) Core(TM) i5-8400 CPU @ 2.80GHz
cpu MHz : 3799.999
cache size : 9216 KB
siblings : 6
cpu cores : 6
My 4 core / 8 thread i7
$ cat /proc/cpuinfo
vendor_id : GenuineIntel
model name : Intel(R) Core(TM) i7-4770 CPU @ 3.40GHz
cpu MHz : 3492.055
cache size : 8192 KB
siblings : 8
cpu cores : 4
--
dik
Rendered 920576 of 921600 pixels (99%)
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
dick balaska <dic### [at] buckosoft com> wrote:
> > Do you have a link? There are some good deals at BB in a "salvage parts
> > from an old PC to help build a new one" sense. I.e. I have an SSD and
> > graphics card already that I could put into one of these machines.
I could fill several 4-foot cubes with all the motherboards, HDD's, memory, etc
that we have here.
Whole pallets of towers, SFF, USDT, laptops, servers, ....
> Pretty weird. This is the box I bought, the link is from my history.
> But, I did *not* pay $800 with monitor. I would have gone for a lesser
> machine.
Let me know if you need anything, and I can see what we've got that is or is not
currently listed. I don't have any control over the pricing or negotiations,
but it's _gotta_ be cheaper than brand new from Best Buy.
And I can hand pick and pack.
http://www.cecent.com/
https://www.ebay.com/usr/cecet
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 08/23/2018 06:15 AM, dick balaska wrote:
> On 08/23/2018 06:07 AM, dick balaska wrote:
>
>>
>>
https://www.bestbuy.com/site/dell-inspiron-desktop-intel-core-i5-12gb-memory-1tb-hard-drive-black-with-silver-trim/6228201.p?skuId=6228201
>>
>
> "Smart six-core, twelve-way processing performance"
> No it doesn't.
>
> $ cat /proc/cpuinfo
> vendor_id : GenuineIntel
> model name : Intel(R) Core(TM) i5-8400 CPU @ 2.80GHz
> cpu MHz : 3799.999
> cache size : 9216 KB
> siblings : 6
> cpu cores : 6
>
>
> My 4 core / 8 thread i7
> $ cat /proc/cpuinfo
> vendor_id : GenuineIntel
> model name : Intel(R) Core(TM) i7-4770 CPU @ 3.40GHz
> cpu MHz : 3492.055
> cache size : 8192 KB
> siblings : 8
> cpu cores : 4
>
I came across an article a while back about Intel dropping
hyper-threading in favor of cores at the lower end. Looks like yours is
one of those processors!
More cores is better for POV-Ray so this no threading more cores change
helps us - and especially in v3.8 over v3.7 given it's threads > cores
performance has slowed more than threads <= cores has (6% vs about 3%
last I did any measures).
Note that your i5-8400 has a turbo frequency of 'up to' 4GHz throttled
by temperature. Many other processors have the turbo option too though
often it's not a much higher frequency than the normal in older
generations. If turbo mode is not off turned off in bios, performance
measures to 'clock time' can be all over the place - better to measure
hardware cycles and even there you need to be sure you are not thrashing
your caches with other work for example.
My little i3-4130 doesn't have the turbo mode which makes timing
relative to the 'clock' a little more reliable.
Well, excepting starting earlier this year when measured time seemed to
get more unstable. Has me wondering about possible impacts for some of
those security patches for metldown and specter exposures - but I've
done no digging. Might be I'm just timing different features of POV-Ray
of late that have different register and cache usage patterns.
Aside: This is the first I've noticed /proc/cpuinfo having bug
information. Mine shows:
bugs : cpu_meltdown spectre_v1 spectre_v2 spec_store_bypass l1tf
Any of those gone on your i5-8400?
Bill P.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 8/23/2018 6:07 AM, dick balaska wrote:
> Pretty weird. This is the box I bought, the link is from my history.
> But, I did *not* pay $800 with monitor. I would have gone for a lesser
> machine.
>
> Technically, the machine is for the 6 year old to play minecraft. :)
>
>
https://www.bestbuy.com/site/dell-inspiron-desktop-intel-core-i5-12gb-memory-1tb-hard-drive-black-with-silver-trim/6228201.p?skuId=6228201
>
This one is slightly cheaper:
https://www.bestbuy.com/site/hp-pavilion-desktop-amd-ryzen-5-series-12gb-memory-1tb-hard-drive-ash-silver/6221040.p?skuId=6221040
It has 4 cores/8 threads, so it might be better for POV-Ray.
Also the APU in the new AMD PCs are supposed to be good. However, I
haven't yet found a way to directly compare this APU with my old GeForce
GTX 750 TI. If the APU is better then I might buy this PC.
Mike
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
dick balaska <dic### [at] buckosoftcom> wrote:
> I've noticed a Ghz decrease in my last couple of boxes. From 3.4Ghz to
> 3.0Ghz and now 2.8Ghz. Certainly that will run "cooler".
> The box is in the living room and is silent, as opposed to my i7 which
> has dual fans that have a teeth-floating harmonic.
Dual fans? Does it overclock? Overclocking is a disqualifier for me; it
increases the risk of frying the CPU without providing any long term advantages
for a ray tracer.
But I can't live without my hyper-threading, so my next box will surely be
another i7.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Am 23.08.2018 um 20:33 schrieb Cousin Ricky:
> But I can't live without my hyper-threading, so my next box will surely be
> another i7.
Not necessarily. IIRC Intel has just recently started to introduce
hyperthreading into the i5 class of CPUs.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 8/23/2018 4:03 PM, clipka wrote:
> Am 23.08.2018 um 20:33 schrieb Cousin Ricky:
>
>> But I can't live without my hyper-threading, so my next box will surely be
>> another i7.
>
> Not necessarily. IIRC Intel has just recently started to introduce
> hyperthreading into the i5 class of CPUs.
>
I thought the i5s always used hyperthreading?
Mike
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 8/23/2018 4:03 PM, clipka wrote:
> Am 23.08.2018 um 20:33 schrieb Cousin Ricky:
>
>> But I can't live without my hyper-threading, so my next box will surely be
>> another i7.
>
> Not necessarily. IIRC Intel has just recently started to introduce
> hyperthreading into the i5 class of CPUs.
>
https://en.wikipedia.org/wiki/List_of_Intel_Core_i5_microprocessors#Westmere_microarchitecture_(1st_generation)
1st generation i5s featured hyperthreading according to Wikipedia.
Mike
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Am 24.08.2018 um 01:48 schrieb Mike Horvath:
> On 8/23/2018 4:03 PM, clipka wrote:
>> Am 23.08.2018 um 20:33 schrieb Cousin Ricky:
>>
>>> But I can't live without my hyper-threading, so my next box will
>>> surely be
>>> another i7.
>>
>> Not necessarily. IIRC Intel has just recently started to introduce
>> hyperthreading into the i5 class of CPUs.
>>
>
> I thought the i5s always used hyperthreading?
From what I'm finding on the internet, you're right in that i5 with
hyperthreading did exist previously, but only in 2-core (physical) models.
With Generation 8(?), Intel has apparently re-introduced HT to i5.
Reliable sources? Can't find any ad-hoc. Would have to dig through a
list of CPUs. No fun, no do.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 08/23/2018 02:33 PM, Cousin Ricky wrote:
> dick balaska <dic### [at] buckosoftcom> wrote:
>> I've noticed a Ghz decrease in my last couple of boxes. From 3.4Ghz to
>> 3.0Ghz and now 2.8Ghz. Certainly that will run "cooler".
>> The box is in the living room and is silent, as opposed to my i7 which
>> has dual fans that have a teeth-floating harmonic.
>
> Dual fans?
Dual fans here means actually three fans. :) One on the power supply,
one on the case, and one on the CPU. All three are different speeds and
make for some irksome high frequency harmonics.
> Does it overclock? Overclocking is a disqualifier for me; it
> increases the risk of frying the CPU without providing any long term advantages
> for a ray tracer.
Any overclocking I do is provided by official intel. I.e. the new i5
overclocks itself. "turbo mode"
model name : Intel(R) Core(TM) i5-8400 CPU @ 2.80GHz
cpu MHz : 3799.999
The thing has been running flat out for a week. I expected it to
slowdown when it got warm.
>
> But I can't live without my hyper-threading, so my next box will surely be
> another i7.
I don't know, man. I like my i7, but this new i5 is *fast*.
I reran a scene and this is the number of frames that each contributed:
name frames % of total
contrib
james 1275 32.93 # i5-8400 @ 2.8GHz (6 cores 6 threads)
cyd 789 20.38 # i7-4770 @ 3.4Ghz (4 cores, 8 threads)
joe 701 18.10 # i5-7400 @ 3.0GHz (4 cores) (2017 box)
ringo 618 15.96 # i5-4460 @ 3.2GHz (4 cores) (2015 box)
liza 489 12.63 # i5-3330 @ 3.0GHz (4 cores) (2013 box)
That scene ran for 8 days and the new i5 was only up for 6 days. I'd bet
the next scene he gets close to 40% of the total work.
--
dik
Rendered 920576 of 921600 pixels (99%)
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 08/23/2018 11:07 AM, William F Pokorny wrote:
> Aside: This is the first I've noticed /proc/cpuinfo having bug
> information. Mine shows:
>
> bugs : cpu_meltdown spectre_v1 spectre_v2 spec_store_bypass l1tf
>
> Any of those gone on your i5-8400?
They are the same set.
The i7 is different:
bugs : cpu_meltdown spectre_v1 spectre_v2 spec_store_bypass
Last years box:
model name : Intel(R) Core(TM) i5-7400 CPU @ 3.00GHz
bugs : cpu_meltdown spectre_v1 spectre_v2 spec_store_bypass
My oldest box is running ubuntu 14 / linux 3.13 (as opposed to linux
4.4) and does not report the bugs.
>
> Bill P.
--
dik
Rendered 920576 of 921600 pixels (99%)
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
dick balaska <dic### [at] buckosoftcom> wrote:
> My oldest box is running ubuntu 14 / linux 3.13 (as opposed to linux
> 4.4) and does not report the bugs.
What's the most memory you have in an individual box?
What have you topped out at, rendering a frame?
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 08/24/2018 05:34 AM, dick balaska wrote:
> On 08/23/2018 02:33 PM, Cousin Ricky wrote:
>
> I don't know, man. I like my i7, but this new i5 is *fast*.
> I reran a scene and this is the number of frames that each contributed:
>
> name frames % of total
> contrib
> james 1275 32.93 # i5-8400 @ 2.8GHz (6 cores 6 threads)
> cyd 789 20.38 # i7-4770 @ 3.4Ghz (4 cores, 8 threads)
> joe 701 18.10 # i5-7400 @ 3.0GHz (4 cores) (2017 box)
> ringo 618 15.96 # i5-4460 @ 3.2GHz (4 cores) (2015 box)
> liza 489 12.63 # i5-3330 @ 3.0GHz (4 cores) (2013 box)
>
> That scene ran for 8 days and the new i5 was only up for 6 days. I'd bet
> the next scene he gets close to 40% of the total work.
>
Interesting - especially it not slowing over time. Supposing:
grep MHz /proc/cpuinfo
shows all six cpu values?
Mike posted a wikipedia link for the i5 generations and something which
caught my eye there is the turbo boost is being applied to your
processor relatively evenly across all cores. 10/10/11/11/11/12
multipliers of 100mhz where on older processors it was more unevenly
boosted and by less. Expect this playing a part in your good performance
too.
I wonder if having the hyperthreading circuitry off in each core - my
understanding is it is there, but disabled - allows each core to run
cooler? There would be less demand for on die core specific storage too
I guess with just one thread.
Christoph, You wondered about articles. Found at least one of the
articles I'd read about more cores instead of threads:
https://arstechnica.com/gadgets/2018/07/leaked-benchmarks-show-intel-is-dropping-hyperthreading-from-i7-chips/
My memory not quite right on things with respect to the current
generation of i5s, but the more cores no threads direction at similar
price is good news for POV-Ray.
Bill P.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Am 24.08.2018 um 15:51 schrieb William F Pokorny:
> Interesting - especially it not slowing over time. Supposing:
Why should they?
Modern CPU cooling systems dissipate heat very quickly. If POV-Ray were
to max out the cooling system's capacity, it would show within a couple
of seconds.
> Mike posted a wikipedia link for the i5 generations and something which
> caught my eye there is the turbo boost is being applied to your
> processor relatively evenly across all cores. 10/10/11/11/11/12
> multipliers of 100mhz where on older processors it was more unevenly
> boosted and by less. Expect this playing a part in your good performance
> too.
>
> I wonder if having the hyperthreading circuitry off in each core - my
> understanding is it is there, but disabled - allows each core to run
> cooler? There would be less demand for on die core specific storage too
> I guess with just one thread.
Probably. There's more chip area than if the HT circuitry were
completely absent, so there's more area to dissipate the heat. Also, the
components that would normally be shared between HT units won't be maxed
out and thus create less heat. So given the same cooling system, with HT
off the CPU will indeed inevitably run cooler. But the CPU may be
specified at a lower TDP, in which case a weaker cooling system may be
employed and thus resulting in the same CPU temperature.
I'm not sure whether Intel does indeed produce cores with HT circuitry
existing yet disabled, but producing non-HT CPUs in this manner does
make sense: You can just produce HT CPUs, and if during die testing you
find that a CPU has defective "semi-cores" you can just blow a fuse to
disable HT and sell it as a lower-tier CPU.
Intel has employed this strategy at least as early as the 80486 days,
when they sold dies with functional FPUs as 80486-DX, and those with
faulty FPUs got a fuse blown and sold as FPU-less 80486-SX. And the
external 80487 FPUs you could buy to upgrade such systems were actually
fully-fledged 80486-DX dies that would just take over the entire system,
disabling the 80486-SX installed in the mainboard's CPU socket.
> My memory not quite right on things with respect to the current
> generation of i5s, but the more cores no threads direction at similar
> price is good news for POV-Ray.
Please don't call it "no threads". Threads are a software thing, a means
of utilizing multiple cores by a single application without too much
overhead. The hardware thing is "hyperthreading", a means of multiple
execution units sharing certain functional blocks in the CPU to utilize
them more efficiently.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 8/24/2018 9:51 AM, William F Pokorny wrote:
> james 1275 32.93 # i5-8400 @ 2.8GHz (6 cores 6 threads)
> cyd 789 20.38 # i7-4770 @ 3.4Ghz (4 cores, 8 threads)
Interesting. I would have expected the bottom CPU to perform better due
to more threads and higher clock speed.
Mike
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Le 24/08/2018 à 20:51, Mike Horvath a écrit :
> On 8/24/2018 9:51 AM, William F Pokorny wrote:
>> james 1275 32.93 # i5-8400 @ 2.8GHz (6 cores 6 threads)
>> cyd 789 20.38 # i7-4770 @ 3.4Ghz (4 cores, 8 threads)
>
> Interesting. I would have expected the bottom CPU to perform better due
> to more threads and higher clock speed.
>
>
> Mike
But nobody looks at the speed of the memory (from tomhardware website):
* i5-8400 uses ddr4 2666
* i7-4770 uses ddr3 1600
DDR4 2666 has a bandwidth of 21333 MB/s
DDR3 1600 has a bandwidth of 12.8 GB/s
12.8 vs 21.3 !
That's the advantage of the new generation.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Am 24.08.2018 um 20:51 schrieb Mike Horvath:
> On 8/24/2018 9:51 AM, William F Pokorny wrote:
>> james 1275 32.93 # i5-8400 @ 2.8GHz (6 cores 6 threads)
>> cyd 789 20.38 # i7-4770 @ 3.4Ghz (4 cores, 8 threads)
>
> Interesting. I would have expected the bottom CPU to perform better due
> to more threads and higher clock speed.
i7-4770 has a max clock of 3.9 GHz,
Despite being rated at 2.8 GHz, the i5-8400 can actually go up to 4.0
GHz, given adequate cooling. The i7-4770 can only go as high as 3.9 GHz.
The i5-8400 also has more cores, which is more important than the number
of concurrent threads. Each thread beyond the number of cores only
translates to about 20% to 25% a core's worth of added computing power,
but only about 1/5 to 1/4 a core's worth, so the i7-4770 is effectively
worth about 5 threads.
In addition, the i5-8400 is four generations newer than the i7-4770,
which presumably translates to some additional performance improvements
here and there from miscellaneous optimizations.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Am 24.08.2018 um 21:39 schrieb Le_Forgeron:
> Le 24/08/2018 à 20:51, Mike Horvath a écrit :
>> On 8/24/2018 9:51 AM, William F Pokorny wrote:
>>> james 1275 32.93 # i5-8400 @ 2.8GHz (6 cores 6 threads)
>>> cyd 789 20.38 # i7-4770 @ 3.4Ghz (4 cores, 8 threads)
>>
>> Interesting. I would have expected the bottom CPU to perform better due
>> to more threads and higher clock speed.
>>
>>
>> Mike
>
> But nobody looks at the speed of the memory (from tomhardware website):
> * i5-8400 uses ddr4 2666
> * i7-4770 uses ddr3 1600
>
> DDR4 2666 has a bandwidth of 21333 MB/s
> DDR3 1600 has a bandwidth of 12.8 GB/s
>
> 12.8 vs 21.3 !
>
> That's the advantage of the new generation.
I'm not sure that translates to much of a performance gain in POV-Ray.
The memory bandwidth limits the peak throughput under sequential read
access.
Most of POV-Ray's memory accesses are probably random access though, and
there memory latency is the limiting factor, not bandwidth.
In terms of latency, DRAM hasn't seen any noticeable improvement for ages.
Also, RAM access doesn't seem to be a bottleneck for POV-Ray anyway:
There are few applications (even computation-heavy ones) that drive CPU
temperatures quite as high as POV-Ray, indicating that the CPU spends a
lot of time doing actual work rather than waiting for data to be fetched
from memory. Presumably POV-Ray's most heavily accessed data typically
fits well within the L3 cache.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
dick balaska <dic### [at] buckosoftcom> wrote:
> Any overclocking I do is provided by official intel. I.e. the new i5
> overclocks itself. "turbo mode"
By "turbo mode," do you mean turbo boost? That is not the same as overclocking.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 08/24/2018 01:58 PM, clipka wrote:
> Am 24.08.2018 um 15:51 schrieb William F Pokorny:
>
>> Interesting - especially it not slowing over time. Supposing:
>
> Why should they?
>
> Modern CPU cooling systems dissipate heat very quickly. If POV-Ray were
> to max out the cooling system's capacity, it would show within a couple
> of seconds.
>
I've not seen that behavior for an extended render on a machine with
turbo boost active is the short answer.
My old i7-920 - before I over-clocked it with the turbo mode turned off
- didn't hold the turbo boost frequency for long. I was doing long
single image isosurface stuff at the time. Dick appears to be getting a
3.8GHz 'core' over time and under load.
A reason I suggested the grep command is I wonder if all the cores are
maxed out with respect to the multipliers. Are some cores in fact being
throttled? The 3.8ghz number is a core seeing a 10x 100Mhz but you can
get up to 12x 100MHz on a core. I don't know if thermal throttling is by
core or by die.
I further wonder how constant the reported 3.8GHz number really is. On
my i3 the frequency values reported in /proc/cpuinfo look to be
relatively instantaneous. I'm not under load and those MHz values change
dramatically every time I look. I think we'd need to monitor the
performance over time for a full image render to be sure the 3.8GHz held.
Dick is rendering frames. Maybe there is in fact some periodicity in the
actual performance where a frame starts with turbo boost maxed out and
then it tails off. Maybe thermal stuff isn't an issue because he is
rendering frames and there is a cool off period between frames.
I worked for a long time on the hardware design side of CPUs and cache
chips and power consumption/heat generation grows exponentially with
frequency increases. Your right, if limits kick in, it happens fast. My
thinking is 2.8GHz is the performance Intel can guarantee under all
loads given the manufacturing process variation and shipped cooling at
some expected "end of life." Otherwise, they'd be claiming a better base
performance & charging for it, or, somehow handicapping the CPUs aimed
at lower end boxes so as not to eat into their higher end - higher
profit - business.
Though, AMD is again a serious player, maybe they are somewhat being
forced to eat the high end business and what we see with Dick's new i5
is a result.
All just me pondering. I don't know anything for sure - even that. :-)
>
> Probably. There's more chip area than if the HT circuitry were
> completely absent, so there's more area to dissipate the heat. Also, the
> components that would normally be shared between HT units won't be maxed
> out and thus create less heat. So given the same cooling system, with HT
> off the CPU will indeed inevitably run cooler. But the CPU may be
> specified at a lower TDP, in which case a weaker cooling system may be
> employed and thus resulting in the same CPU temperature.
>
> I'm not sure whether Intel does indeed produce cores with HT circuitry
> existing yet disabled, but producing non-HT CPUs in this manner does
> make sense: You can just produce HT CPUs, and if during die testing you
> find that a CPU has defective "semi-cores" you can just blow a fuse to
> disable HT and sell it as a lower-tier CPU.
>
> Intel has employed this strategy at least as early as the 80486 days,
> when they sold dies with functional FPUs as 80486-DX, and those with
> faulty FPUs got a fuse blown and sold as FPU-less 80486-SX. And the
> external 80487 FPUs you could buy to upgrade such systems were actually
> fully-fledged 80486-DX dies that would just take over the entire system,
> disabling the 80486-SX installed in the mainboard's CPU socket.
>
Agree & that last bit about the 80487's being in fact 80486-DX dies I
did not know.
>
>> My memory not quite right on things with respect to the current
>> generation of i5s, but the more cores no threads direction at similar
>> price is good news for POV-Ray.
>
> Please don't call it "no threads". Threads are a software thing, a means
> of utilizing multiple cores by a single application without too much
> overhead. The hardware thing is "hyperthreading", a means of multiple
> execution units sharing certain functional blocks in the CPU to utilize
> them more efficiently.
>
Good point. I got sloppy with the wording.
Bill P.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Am 25.08.2018 um 18:10 schrieb William F Pokorny:
> A reason I suggested the grep command is I wonder if all the cores are
> maxed out with respect to the multipliers. Are some cores in fact being
> throttled? The 3.8ghz number is a core seeing a 10x 100Mhz but you can
> get up to 12x 100MHz on a core. I don't know if thermal throttling is by
> core or by die.
Where do you get those numbers from?
From all I know (and what a quick random peek at the internet
confirmed), CPU speed is per core, not per package.
> I further wonder how constant the reported 3.8GHz number really is. On
> my i3 the frequency values reported in /proc/cpuinfo look to be
> relatively instantaneous. I'm not under load and those MHz values change
> dramatically every time I look. I think we'd need to monitor the
> performance over time for a full image render to be sure the 3.8GHz held.
What you might be seeing is the CPU not throttling due to overheating,
but due to being bored.
Also, if this is a per-CPU speed, it may be averaging over all cores,
even cores shut down entirely due to being bored.
> My thinking is 2.8GHz is the performance Intel can guarantee under all
> loads given the manufacturing process variation and shipped cooling at
> some expected "end of life." Otherwise, they'd be claiming a better base
> performance & charging for it, or, somehow handicapping the CPUs aimed
> at lower end boxes so as not to eat into their higher end - higher
> profit - business.
Certainly. At the miniscule dimensions they're currently working at, I'd
be surprised if their tests could catch all the bad apples; so it's
reasonable to assume that they're working with safety margins, and
rating the CPUs for the worst case scenario.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 08/25/2018 01:26 PM, clipka wrote:
> Am 25.08.2018 um 18:10 schrieb William F Pokorny:
>
...
>
> Where do you get those numbers from?
They came off the Wikipedia page Mike posted in this thread.
>
> From all I know (and what a quick random peek at the internet
> confirmed), CPU speed is per core, not per package.
>
>
...
>
> What you might be seeing is the CPU not throttling due to overheating,
> but due to being bored.
>
> Also, if this is a per-CPU speed, it may be averaging over all cores,
> even cores shut down entirely due to being bored.
>
I wasn't trying to suggest overheating on my machine. My machine is not
under load - just thunderbird and some xterm windows up. The point I was
trying to make is how quickly the frequency numbers in /proc/cpuinfo
change. Suppose Dick's machine starts a new frame. At the very beginning
things are cool. The CPU (I suspect rather each core as you suggest)
detects it has meaningful stuff to do so turbo boost it is. Initially no
reason not to max turbo boost out; things are cool. Suppose it's in
those first few seconds Dick grabs the frequency values from
/proc/cpuinfo. They're going to be maxed out. Maybe those maxed out
turbo frequencies hold for the entire frame, maybe not. We don't really
know unless we sample at a fairly high rate over time.
Otherwise, I agree my machine is part of the time idling at a lower
frequency and very likely operating voltage too. I see quite large
differences between the 4 hyper-threads over time with a bottom at
800MHz and a top end of 3400MHz, but I've got no idea how the detailed
accounting works. I expected the hyper-threads to hang together in pairs
with respect to frequency as there are two per core, but it's not
reported that way.
>
...
>
A couple thoughts I had this afternoon along the lines of your thinking
there is plenty of cooling to hold the max turbo frequencies. Supposing
the i5s are aimed mostly at laptops; the cooling in a box enclosure is
much better. Second, my i7 920 experience with turbo boost is 10 years
dated.
Bill P.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 08/23/2018 06:15 AM, dick balaska wrote:
> $ cat /proc/cpuinfo
> model name : Intel(R) Core(TM) i5-8400 CPU @ 2.80GHz
> siblings : 6
> cpu cores : 6
I'll bet the argument on linux-kernel was fun.
"Dude, it's 5 siblings!"
Linus: "Bite me. I say 6."
(Linus is a potty mouth)
--
dik
Rendered 1024 of 921600 pixels (0%)
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
