[HN Gopher] Core I9 14900KF Breaks World Record, Almost Achieves...
___________________________________________________________________
Core I9 14900KF Breaks World Record, Almost Achieves 9.1GHz (2023)
Author : EveryPizza
Score : 91 points
Date : 2024-03-24 01:39 UTC (21 hours ago)
(HTM) web link (www.tomshardware.com)
(TXT) w3m dump (www.tomshardware.com)
| gnabgib wrote:
| October 2023... and that typo.. The team was able
| to achieve a very impressive 9043.92GHz
| m463 wrote:
| if you want to properly make light of it, you have to say
| 480000Ghz
|
| (a literal joke, since visible light is 480 thz)
| sourcecodeplz wrote:
| Dunno what happened but way in the day tomshardware writers
| were very knowledgeable and quite witty + smart too. Guess it
| got sold? I can't pronounce the name of more than 80% of their
| writers now.
| fxtentacle wrote:
| The writer in question seems to be a freelancer who started
| working with them in August 2023, meaning he was pretty new
| to Tomshardware when he published this in October.
|
| Looks like the usual cost cutting where you replace employees
| with contractors.
|
| Also, this entire article is based on an Asus advertisement
| video on YouTube. I'm sure they wouldn't put their best
| writers on that kind of content.
| AlecSchueler wrote:
| > I can't pronounce the name of more than 80% of their
| writers now.
|
| What does this mean? Only those with Anglo-Saxon heritage
| make for good writers?
| BenjiWiebe wrote:
| Anglo-saxon names would make it more likely that English is
| their first language (and therefore more likely to be
| fluent), which is the language used in these articles.
|
| A bit of a stretch though.
| AlecSchueler wrote:
| I'm not sure that's true.
|
| Firstly because countries like the US are home to many
| native English speaking people with non-Saxon names, and
| countries like Ireland in the old world speak English
| while retaining their native names.
|
| Secondly because many people who learn English as a
| second language grow up be fluent and can write with
| ease. I look to places like the Netherlands where
| students from all over the world are writing theses for
| their higher education completely in English.
|
| It's not that hard to write in your second language,
| certainly not to the point that your name could be in any
| way an indicator of your skill level.
| ithkuil wrote:
| It's certainly easier to write well in your second
| language than to speak it well!
| underlogic wrote:
| Waste of sand. You want liquid nitrogen on your desk? Where's the
| utility?
| htag wrote:
| It's liquid helium, which I imagine just evaporates into the
| air.
| bee_rider wrote:
| Although, wouldn't liquid nitrogen also evaporate? The
| atmosphere has lots of nitrogen.
| olliej wrote:
| The bigger issue is that helium is a functionally finite
| resource, and does not remain in the atmosphere (at least
| IIRC it literally just floats up until it gets blown into
| space).
|
| The bigger local issue you run into with liquid helium and
| liquid nitrogen is having it evaporate in an enclosed
| space. You can "easily" create an environment leading to
| inert gas suffocation (a real hazard in some industrial
| cases) - in reality any simple case like this is unlikely
| to be using enough of N or He, and is unlikely to be
| sufficiently enclosed, but in principle it would be
| possible - maybe if you were in a basement and spilled an
| inexplicably large thermos of them it could do it.
| lathiat wrote:
| You can definitely cause problems by spilling an entire
| dewer into a smaller room.
|
| Had to calculate this for the safety part of a LAN party
| event application. Fortunately we had a very big room.
| olliej wrote:
| Dewer! thanks, I knew there was an actual name for those
| heavy duty thermoses but it completely escaped me :D
| BenjiWiebe wrote:
| Dewar, actually. Originally Dewar flask, after Sir James
| Dewar.
| apantel wrote:
| It's a sport. It's like drag racing. Nobody is going to drive a
| drag racer to work, but that's not the point of the car.
| arcanemachiner wrote:
| Waste of cars. You want liquid nitromethane on your roads?
| Where's the utility?
| olliej wrote:
| liquid nitromethane commuting? in this economy? :D
| oorza wrote:
| "In this economy" snark? In this economy?
| rgmerk wrote:
| But at least drag racing is a spectacle. Not my thing, but it
| is a spectacle.
|
| This? You push a button and a number comes up.
| lostlogin wrote:
| Benchmarking hardware has been a thing forever, and applies
| to a vast swathe of things. Graphics cards, ram, hard
| drives, network cards etc.
|
| Then you look further and see that it's done with coffee
| machines, motors (of all sorts), and just about any other
| device you can find, make or name.
|
| Wanting a fast/strong/powerful/quick X is a fairly common
| thing for many of us.
| steve_rambo wrote:
| Are you sure you haven't visited _hacker_ news by mistake?
|
| Hardware overclocking requires a decent amount of
| knowledge, most of it obtainable only through months of
| trial and error, and a lot of tuning to push dozens of
| often conflicting parameters just right. Extreme
| overclocking requires that much more. If what they're doing
| is simply "pushing a button", then programming and system
| administration can be reduced to that too, along with many
| other things.
|
| You simply can't get full performance from modern systems
| without some amount of overclocking, and things like PBO
| and XMP/EXPO profiles are far from what your hardware can
| achieve because they have to be very conservative, or many
| systems won't run without additional manual tuning, which
| most consumers won't do.
|
| (Except for closed systems like Apple's where your hardware
| doesn't belong to you and you can't change anything
| anyway.)
|
| So one immediate thing overclockers provide are general
| guidance on what you can expect to achieve and what thing
| to tune which way to get close to maximum performance from
| your hardware without spending months on it like they did.
| I heavily rely on such information. My system would be at
| least 25% slower if not for these "button pushers".
| underlogic wrote:
| I wouldn't trust an overclocked system as a server or to
| compile. Especially not long term w degradation. Can't
| use it for CAD bc it'll mess up the PCIe timings I think.
| Looks like a giant waste of time and money to me, but
| sure to each their own. I bet it's difficult, but seems
| also pointless. If they were really knowledgeable they'd
| get out an FPGA and go design their own SoC instead of
| filing down intel's for a speed bump. Just not relevant
| today
| tomohelix wrote:
| Look to me a significant amount of engineering went into making
| sure the chip can withstand the amount of power it is drawing.
| Basically a desperate move by Intel to keep their outdated design
| from being completely smoked by AMD.
| fortran77 wrote:
| Intel isn't worried about AMD. They're worried about Arm.
| arcticbull wrote:
| Intel is getting attacked from all sides.
|
| AMD has raised their market share over the last 4-5 years
| from about 8% to 31% of x86 sales. Intel also saw 5 straight
| years of market share declines against AMD in the server
| space - which is by far the most lucrative.
|
| And yes, they're _also_ worried about ARM and Nvidia.
| nolok wrote:
| More than half of Intel's revenue come from desktop and
| laptop chips, and that's the segment that's being eaten the
| fastest, and the one doing the eating is AMD.
|
| ARM is another threat on the horizon on that front, but it's
| nothing compared to the beating AMD has been giving them
| since the Ryzen showed up.
| aurareturn wrote:
| It's true. It's hard to see how Intel can catch up to Apple
| and AMD in the client and server space respectively.
|
| However, I still own a substantial amount of Intel stocks
| because it will be one of the most valuable company in the
| world if China uses military action against Taiwan. If not,
| then I still believe Intel will be successful with their
| IFS strategy because the world is about to enter an era
| where compute will never be enough and designers want a
| second supplier next to TSMC.
| Aromasin wrote:
| I'm also backing that investment thesis. I don't have
| hopes for Intel being anything but a budget alternative
| in terms of their client and datacentre products, but the
| foundry business is only going to grow as more players
| join the gold rush that is compute power, and they're
| going to be the ones selling the picks.
| Detrytus wrote:
| Well, give it 1-2 more years and they will leapfrog TSMC
| with their manufacturing process. 18A is allegedly still
| on track, and Intel is also getting into foundry
| business, I can totally imagine NVIDIA switching to Intel
| for their next-gen GPUs for example. That, combined with
| worries about China attacking Taiwan makes me bullish on
| Intel stock.
| aurareturn wrote:
| I doubt they leap frog TSMC. 18A will come out around the
| same time as N2. I have more confidence that TSMc's N2
| will perform better and be in far higher volume and yield
| rate.
|
| 20A is sort of Intel's first foray. 18A is really their
| serious node.
|
| If you think Intel has a chance to leap TSMC, it will
| come at the 14A node but you have to bet that TSMC is
| wrong to stay with low NA.
| formerly_proven wrote:
| Their process strategies lined up at final gen DUV and
| 1st gen EUV, but are divergent for the future: TSMC isn't
| using high NA EUV for the time being and instead focuses
| on multi-patterning. Intel is instead trying to push high
| NA EUV into production first. We will see which works
| out; they can of course also both achieve success in
| different ways.
| aurareturn wrote:
| Given the recent history, it's safer to bet that TSMC
| knows what they're doing more. It feels like Intel is
| risking a lot to try to win back the crown. Hence, I
| believe in TSMC's more.
| Detrytus wrote:
| Given the recent history TSMC is kind of repeating
| Intel's mistake here: they do not want to pay for new
| generation of the ASML litography machines (which are
| like $300M a piece), so they will leverage the last-gen
| ones and try multi-patterning instead. Didn't work for
| Intel last time, will end badly for TSMC this time.
| formerly_proven wrote:
| That seems to be the precedent for this exact situation -
| iirc both the slight Intel 14nm and the large Intel 10nm
| delays were due to getting multi-patterning working at
| scale.
| imtringued wrote:
| We already live in an era of compute saturation for
| machine learning. The NPUs that AMD and Intel ship have
| more compute than the memory bus can handle.
| aurareturn wrote:
| Explain more? Are you saying we have enough compute but
| not enough memory?
| nolok wrote:
| Much better than the time just a few years ago where they tried
| to show a 28 core 5ghz chip and pretend it was a normal one and
| got pissed when enthousiast realized they were using insane
| cooling to get it there.
|
| I'm still baffled by that showing years later. Over-engineered,
| over-cooled chips to reach absurd speed record has been a
| staple since as far as I remember, like back in the Pentium 2
| or before. Why did anyone at Intel think they should hide the
| sauce, or get pissed when fans got to it, is beyond my
| comprehension.
| PartiallyTyped wrote:
| > enthousiast realized they were using insane cooling to get
| it there.
|
| Didn't they use a chiller?
| nolok wrote:
| They used a 1700 watt chiller with sub ambient cooling, and
| a power unit who could pump up to 2000 watts into the CPU.
| So clearly "record for the show", not "here is what you
| will buy tomorrow".
|
| Which, taken as the usual world record and cool feat of cpu
| speed would have been perfectly fine and impressive, but
| they were getting scared of AMD so for some weird reason
| some idiot in their PR team insisted on pretending from top
| to bottom that this was a normal chip running on normal
| condition, and when Asus (I think ?) let fans see how it
| was achieved Intel gave them less than 30 minutes to give
| the chip back.
|
| This was not a great era for Intel ...
| NoPicklez wrote:
| Not at all really.
|
| The 9Ghz clock was achieved not through any normal cooling or
| by efficiency of the chip.
|
| These overclocking records have been around for decades but
| they're in no way shape or form representative of the average
| of even the top 1% of users.
|
| It's impressive purely because it was possible with an off the
| shelf chip.
| Klaster_1 wrote:
| Obligatory soundtrack:
| https://www.youtube.com/watch?v=u35crz9xZG4
| bee_rider wrote:
| I'll always wonder what would have happened if Intel had stuck
| with Pentium 4 and those incredibly long (31 stages! How we've
| fallen). Sure, they were power hungry, but at ~100W they don't
| compare badly to modern chips. And dumping 100 Watts onto a
| single core is an extremely cool and fun thing to do. I wonder if
| we could have 10GHz processors for real by now.
| arcticbull wrote:
| Sorry, why do you say that the P4/Netburst microarchitecture
| doesn't compare badly to modern chips? Their performance was
| utter garbage (at the time, and now) which is why the Pentium M
| architecture (a Pentium III derivative) was used when they
| built out Core. AMD was spanking them at the time with the
| K8/Athlon 64 and the Athlon 64 X2.
|
| A super long pipeline allows higher clock rates but it takes a
| giant dirt nap when branch prediction fails and when you have a
| cache miss. You end up having massive latencies in these cases.
|
| Further, generally all else being equal a lower clock rate
| allows you to be more energy efficient.
| bee_rider wrote:
| Just in terms of power consumption they don't compare badly.
| Performance, of course, no comparison, haha.
|
| Agree that it had tons of problems. But branch prediction has
| gotten better, compilers have gotten better, etc. Maybe they
| could be handled now!
| smolder wrote:
| Anything salvageable from the architecture likely has been
| used again by this point.
| bcrl wrote:
| The worst part of the P4 was when it went off into la-la land
| for 4000 cycles of replay. Did you issue a locked
| instruction? 4000 cycle penalty. Use rep ; movs? 4000 cycle
| penalty. Some weird internal condition with a misaligned
| store? 4000 cycle penalty. One could optimize to improve
| performance in those cases, but the performance glass jaws on
| the P4 were all consuming.
| xgkickt wrote:
| (Edit: With modern core counts) I think we'd be seeing
| partnerships with GE, Whirlpool, Rheem etc ;)
| bee_rider wrote:
| Exactly! We barely interact with our dishwashers,
| refrigerators, and water heaters. We just fill them up and
| empty them out occasionally. Boring! But we're willing to
| devote hundreds or thousands of watts to them and dozens of
| square feet of floor space to them. Computers are much more
| interesting, therefore they deserve that same appliance
| treatment.
|
| I'm kidding in these sense that I don't think a single core
| could be designed to usefully use 1000W. I get why things
| happened as they did. But I do still think single threaded
| performance is much more interesting than multi-core, so I
| wish we could see how those designs would have evolved.
| xgkickt wrote:
| There's so much waste heat that could be diverted into hot
| water storage.
| gpderetta wrote:
| Time for a custom water cooling loop!
| bee_rider wrote:
| At the time you'd start up your computer, then go make
| some coffee or tea while it booted. Imagine if that could
| be combined into one process!
| 0x1ceb00da wrote:
| Obligatory "it's over 9000!"
|
| https://youtu.be/SiMHTK15Pik?si=0wlPOl7QGVf1zTe8
| nokeya wrote:
| In-process soundtrack:
| https://youtu.be/hyNu5i_6lKA?si=fVzH88KRo55L-uG_
| leshokunin wrote:
| Sick overclock. It's cool to see Intel chips reach such new
| heights. It felt like we were stuck for 10 years. Hopefully as
| they get better at manufacturing and shrinking the 5GHz bar can
| be reliably passed.
| formerly_proven wrote:
| 5 GHz in off the shelf SKUs was passed years ago.
| leshokunin wrote:
| Key word is reliably. I don't think we're quite at the stage
| where the average cpu can sustain 5ghz, especially for longer
| periods of time?
| Gravityloss wrote:
| One would think from far away perspective or first principles,
| that we have passed the point of liquid cooling being way more
| optimal in the datacenter a long time ago already (if you're
| actually doing computing and not idling).
|
| But everything's so slow and path dependent.
|
| I wonder how much you could do with a single rack if you got
| really serious about it. Cooling, power, networking etc.
___________________________________________________________________
(page generated 2024-03-24 23:01 UTC)