Intel’s 12-Core i9 CPU May Clock Well below AMD’s Threadripper

Megalith

Megalith

24-bit/48kHz
Staff member
Joined
Aug 20, 2006
Messages
13,000
Specifications included in a leaked price list suggest that Intel’s 12-core i9-7920x has lost the megahertz war to AMD: the $800 12-core Ryzen Threadripper 1920X is clocked at 3.5GHz, while the $1189 i9-7920x only runs at 2.9GHz. Time will tell how these chips truly compare (i.e., how they hold), however. You can see how well the 7900X clocked for HardOCP in this article.

Intel’s price list typically lists the base clock or minimum clock speeds a CPU will run at under the heaviest workloads. In this case, the Core i9-7920X appears to guarantee an all-core clock speed that is 500MHz lower than AMD’s published specs. AMD recently said it will launch its $1,000 16-core Ryzen Threadripper 1950X at 3.4GHz, and its $800 12-core Ryzen Threadripper 1920X at 3.5GHz. On paper that doesn’t look promising for Intel.
 
Last edited by a moderator:
Good.Jpg

I want price wars. I want solid competition. I want AMD to succeed if only to force intel to Innovate and break into new ground. Intel CAN do (and has done) this already if evident but the [H] post of the best mobo's of all time being 100% intel based.

I want intel to throw down the 64 core monster undercutting AMD by 1/2 and say "WHAT NOW MOFUKA!?!?!"

This only benefits us by Intel looking bad and doing poorly. Their stock will plummet, the CEO will get fired, and some dick swinging hard ass will come in with a Steve Jobs mentality of "MAKE THIS SHIT FASTER BETTER CHEAPER OR Y'ALL FIRED"

Please see the Alec Baldwin GlenGary Glen Ross video for example.
 
Bigdady92 said:
Please see the Alec Baldwin GlenGary Glen Ross video for example.
Click to expand...

No one is disputing that AMD finally won themselves a "set of steak knives", but Intel can just throw mountains of money at the problem and win by default.
 
Bigdady92 said:
Good.Jpg

I want price wars. I want solid competition. I want AMD to succeed if only to force intel to Innovate and break into new ground. Intel CAN do (and has done) this already if evident but the [H] post of the best mobo's of all time being 100% intel based.

I want intel to throw down the 64 core monster undercutting AMD by 1/2 and say "WHAT NOW MOFUKA!?!?!"

This only benefits us by Intel looking bad and doing poorly. Their stock will plummet, the CEO will get fired, and some dick swinging hard ass will come in with a Steve Jobs mentality of "MAKE THIS SHIT FASTER BETTER CHEAPER OR Y'ALL FIRED"

Please see the Alec Baldwin GlenGary Glen Ross video for example.
Click to expand...
In intel's defense, there is a huge push for profits and dividends. Investing money in technology is only worth it if there's competition you need to beat.
If Intel's CEO decided to tell it's shareholders that they were reinvesting profits into desktop technologies, he probably would have been fired because there was no competition to warrant that. Instead they've been investing money directly into efficiency to try and get a foothold on mobile chips (which hasn't panned out all that much).
AMD's new chips are like a sucker punch that they laughed off. Now they're not laughing and will probably downplay AMD's success through marketing.
 
horrorshow said:
No one is disputing that AMD finally won themselves a "set of steak knives", but Intel can just throw mountains of money at the problem and win by default.
Click to expand...

That's BILLIONS of dollars of cash in the bank they will have to toss at the 'problem' . With 3rd party vendors finally onboard (Dell, HP, Supermicro) making high end servers that will be significantly cheaper and run faster than Intel the piggy bank is going to be a floodgate of cash coming out.

Good.
 
Intels Salty-ness on their recent press slide thingy was quite hilarious. I think Linus did a video about it. Intel is uncharacteristically straight shit talking AMD. I never thought Id be eating popcorn while watching intel+amd go back and forth.
 
Hypergreatthing said:
In intel's defense, there is a huge push for profits and dividends. Investing money in technology is only worth it if there's competition you need to beat.
If Intel's CEO decided to tell it's shareholders that they were reinvesting profits into desktop technologies, he probably would have been fired because there was no competition to warrant that. Instead they've been investing money directly into efficiency to try and get a foothold on mobile chips (which hasn't panned out all that much).
AMD's new chips are like a sucker punch that they laughed off. Now they're not laughing and will probably downplay AMD's success through marketing.
Click to expand...
Yeah, I think AMDs underestimation by others was near universal.
 
horrorshow said:
No one is disputing that AMD finally won themselves a "set of steak knives", but Intel can just throw mountains of money at the problem and win by default.
Click to expand...
Throwing cash at a "problem" rarely works. It takes true innovation, something Intel has been short on for the last several years. I am not sure if the gargantuan corporation that Intel has become, with entrenched profit margin requirements, will quickly be able to change course. I'm not saying that Intel is headed for a catastrophic encounter with an iceberg, but simply that they will not be able to quickly respond to AMD's innovation, losing a bit of their dominance. Don't expect miracle products from Intel, rather expect a huge push to "encourage" customers to stay with Intel rather than switch to AMD.
 
Dan said:
Intels Salty-ness on their recent press slide thingy was quite hilarious. I think Linus did a video about it. Intel is uncharacteristically straight shit talking AMD. I never thought Id be eating popcorn while watching intel+amd go back and forth.
Click to expand...

You know you're making competitors sweat when they go out of their way to attack you. The worst fate is to be ignored--because it means your competition literally doesn't care what you do because you matter so little to them. Supposedly, Intel's whole CPU business plan assumed "AMD will always suck" which is part of the reason Ryzen caught them on the crapper.
 
At this point if I was building or upgrading I’d likely go with AMD as much to spite Intel and support a competitor even if I lost 10% performance at the same price.

The one issue I’m noticing which I hope changes is motherboard availability. There are 15-20 motherboards from MSI for Z270 and there are 6 for X370. AM4 ITX is almost non-existent.
 
jardows said:
It takes true innovation, something Intel has been short on for the last several years.
Click to expand...

Perhaps they merely haven't shown their hand?

Why feed someone filet mignon when they're perfectly content with flank steak?
 
horrorshow said:
Perhaps they merely haven't shown their hand?

Why feed someone filet mignon when they're perfectly content with flank steak?
Click to expand...
That's not how it works. Intel doesn't have a secret vault of innovations that they dole out based on market share and competition.
 
Dan said:
Intels Salty-ness on their recent press slide thingy was quite hilarious. I think Linus did a video about it. Intel is uncharacteristically straight shit talking AMD. I never thought Id be eating popcorn while watching intel+amd go back and forth.
Click to expand...


Brings back those good ol' days of the Athlon XP era, doesn't it? Its fun.
 
Hypergreatthing said:
That's not how it works. Intel doesn't have a secret vault of innovations that they dole out based on market share and competition.
Click to expand...
Considering that anything in the release pipeline over the course of the next year has been in development for years(and that would apply to Intel, AMD, and Nvidia), it may very well be the case. It's entirely possible for Intel to be sitting on something and not bothering to push it to market due to the investment cost and lack of "need" at the time. With actual competition capable of stealing marketshare in the CPU space now, Intel may actually have to do something rather than just get away with bullshit minor iterations.
 
Rvenger said:
Brings back those good ol' days of the Athlon XP era, doesn't it? Its fun.
Click to expand...
Yup, Im all about whats the best bang for your buck. I dont care who offers it. I was in a position where I was able to upgrade my 3570k. My friend took my Z77 platform off my hands for cheap and I scooped up the 1700. Yea the 7700k is similarly priced but i mean... 8c16t at 3.9ghz is plenty for me. My intentions were not to upgrade the PC for a few years. I think I accomplished my goal
 
But but but..... Didn't Intel say it was all about the gtx? Until they said it wasn't. Until they said it was. Until it wasn't. I am confused. Does this matter? Or was it AMD that said that.....
 
weren't xeon chips always known for lower clocks than their desktop counter parts?

this seems like a given really.
 
Dan said:
Yup, Im all about whats the best bang for your buck. I dont care who offers it. I was in a position where I was able to upgrade my 3570k. My friend took my Z77 platform off my hands for cheap and I scooped up the 1700. Yea the 7700k is similarly priced but i mean... 8c16t at 3.9ghz is plenty for me. My intentions were not to upgrade the PC for a few years. I think I accomplished my goal
Click to expand...


The extra cores/threads in productivity is what sold me. The gaming is on par in my opinion and does not slow down my 1080ti in 4k. I decided I am permanently on 4k now anyways.
 
Design1stcode2nd said:
At this point if I was building or upgrading I’d likely go with AMD as much to spite Intel and support a competitor even if I lost 10% performance at the same price.

The one issue I’m noticing which I hope changes is motherboard availability. There are 15-20 motherboards from MSI for Z270 and there are 6 for X370. AM4 ITX is almost non-existent.
Click to expand...

That is strictly a function of demand in the marketplace. Make no mistake, if Ryzen and Threadripper take off the way I think they will, there will be PLENTY of motherboards to choose from next year!
 
Master_shake_ said:
weren't xeon chips always known for lower clocks than their desktop counter parts?

this seems like a given really.
Click to expand...

Intel will probably just throw more cores and lower cost to compete.

I'm more interested higher clock speeds, not more cores on some of my servers since Microsoft (SQL) is now licensed per core. Anything more than the base 8 cores increases the cost of the software.

It's similar with desktop/laptops.
Most software doesn't even make good use 4 cores, let alone 8 or 16. For office use, I'd rather have a higher clocked 4 core CPU than a lower clock 8 core.
 
Now all we need is for AMD to say "our systems don't have backdoors", and then with any luck we can see Intel ME be 100% optional.

For those unfamiliar, check out the libreboot page on Intel ME, it's fucking scary!
 
Merc1138 said:
Considering that anything in the release pipeline over the course of the next year has been in development for years(and that would apply to Intel, AMD, and Nvidia), it may very well be the case. It's entirely possible for Intel to be sitting on something and not bothering to push it to market due to the investment cost and lack of "need" at the time. With actual competition capable of stealing marketshare in the CPU space now, Intel may actually have to do something rather than just get away with bullshit minor iterations.
Click to expand...
Show me a case where a major player in computer tech has been sitting on secret tech, and revealing it at the last moment in response to what their competitor just did. I can't think of any examples, and seriously doubt it is happening now. Of course, secret, unknown tech will always exist until we have proof that it doesn't!
 
jardows said:
Show me a case where a major player in computer tech has been sitting on secret tech, and revealing it at the last moment in response to what their competitor just did. I can't think of any examples, and seriously doubt it is happening now. Of course, secret, unknown tech will always exist until we have proof that it doesn't!
Click to expand...

So, you're implying, for example, Yum! brands doesn't have a hundred new tasty treats laying in wait, once your eye starts wandering?
 
horrorshow said:
So, you're implying, for example, Yum! brands doesn't have a hundred new tasty treats laying in wait, once your eye starts wandering?
Click to expand...

When Intel were getting clobbered by AMD back in the old days, where was that secret tech then? Or were they just sitting tight waiting for AMD to play itself out? Or were they too busy cutting behind closed door deals to freeze AMD out of the market? Hmm....???
 
Merc1138 said:
Considering that anything in the release pipeline over the course of the next year has been in development for years(and that would apply to Intel, AMD, and Nvidia), it may very well be the case. It's entirely possible for Intel to be sitting on something and not bothering to push it to market due to the investment cost and lack of "need" at the time. With actual competition capable of stealing marketshare in the CPU space now, Intel may actually have to do something rather than just get away with bullshit minor iterations.
Click to expand...
Eh... there's reasons. There may be ideas but due to time constraints, time and money to optimize, or even creating a proof of concept and test platform they were never implemented. That doesn't necessarily mean that they would be effective.
I'll point to the impending vega flop and all the new tech they touted for it as an example.
 
Intel has a long product map of when they will introduce product A, B, C. This is of course based on a number of factors only one of which is research/design. If you don't have any real competition you can cadence your releases to maximize your profits offering X% of increased performance. Now that AMD has appeared to offer competing product they now have to up the cadence and or adjust pricing to maintain or win back market share.

You don't really think Intel is putting out the best they can design/manufacture as fast as they can do you?
 
Design1stcode2nd said:
Intel has a long product map of when they will introduce product A, B, C. This is of course based on a number of factors only one of which is research/design. If you don't have any real competition you can cadence your releases to maximize your profits offering X% of increased performance. Now that AMD has appeared to offer competing product they now have to up the cadence and or adjust pricing to maintain or win back market share.

You don't really think Intel is putting out the best they can design/manufacture as fast as they can do you?
Click to expand...

Yet there are tons of people still on Sandy bridge, how is this slight increase in performance making intel heaps of money? They would make slightly more money if everyone felt the need to upgrade every new generation of CPU's then every 5 or 6 and then mostly for some new features or a hardware failure.
 
horrorshow said:
So, you're implying, for example, Yum! brands doesn't have a hundred new tasty treats laying in wait, once your eye starts wandering?
Click to expand...
My statement, and this discussion, is as it relates to computer technology, not food. Intel, and most all of the major players, have a roadmap of the technology they are working on for the next several years. They have to, since they can't just spring new tech on the industry and have mass adoption. System integrators, OEM's, developers, programmers, and all the like, have to make sure they are ready for the new technology, and that can often take years. In this industry, to hold back on technology that you have invested millions if not billions of dollars in, just to release it at the "right" time to stave off competition, is foolish at best, and suicidal at worst.
 
Denpepe said:
Yet there are tons of people still on Sandy bridge, how is this slight increase in performance making intel heaps of money? They would make slightly more money if everyone felt the need to upgrade every new generation of CPU's then every 5 or 6 and then mostly for some new features or a hardware failure.
Click to expand...

That's what the beancounters are for. If performance increase = X Y number of expected sales. If performance = Z Y number of expected sales. If 2 or 3 cycles of X equals more profit than X then there is no need to just go to Z.

Basically they make more money on the people willing to upgrade to the latest and greatest or just entering the PC market than the extra they would get from the one time larger performance jump enticing people on older platforms to upgrade.
 
jardows said:
Show me a case where a major player in computer tech has been sitting on secret tech, and revealing it at the last moment in response to what their competitor just did. I can't think of any examples, and seriously doubt it is happening now. Of course, secret, unknown tech will always exist until we have proof that it doesn't!
Click to expand...
Wtf are you getting "last minute" from? I even said in plain English this shit takes years to get to market. No one is saying Intel will be able to respond properly to and in 6-12 months, this isn't something I would expect Intel to compete with for 1.5-2 years. Just because these companies can get a product out the door every 6-18 months does not mean that is the development cycle.
 
jardows said:
My statement, and this discussion, is as it relates to computer technology, not food. Intel, and most all of the major players, have a roadmap of the technology they are working on for the next several years. They have to, since they can't just spring new tech on the industry and have mass adoption. System integrators, OEM's, developers, programmers, and all the like, have to make sure they are ready for the new technology, and that can often take years. In this industry, to hold back on technology that you have invested millions if not billions of dollars in, just to release it at the "right" time to stave off competition, is foolish at best, and suicidal at worst.
Click to expand...

It's called an "analogy".

School, much?
 
I hope AMD drove armored cars full of cash to Jim Keller's house. That dude may have single handedly* pulled AMD's burning body out of the dumpster fire they called home.

* yeah yeah, team effort and all that. But without Jim, would there even have been a Ryzen? Or likely just more re-released versions of 'Dozer.

Intel has been too compliant for far too long. Now we see how long it takes them (and how many BOD - Billions Of Dollars) to come up with something new and innovative.
 
Design1stcode2nd said:
Intel has a long product map of when they will introduce product A, B, C. This is of course based on a number of factors only one of which is research/design. If you don't have any real competition you can cadence your releases to maximize your profits offering X% of increased performance. Now that AMD has appeared to offer competing product they now have to up the cadence and or adjust pricing to maintain or win back market share.

You don't really think Intel is putting out the best they can design/manufacture as fast as they can do you?
Click to expand...

Agreed!

Intel has that tech - It just didn't do them any good to compete with themselves - so they had no reason to release it. Now that they have some competition - they can move forward with stuff they've kept on the shelves at the ready.

Take a lesson from Sergey Bubka.

He beat the world record for pole vault 35 times. He'd intentionally beat his own world record by slim amounts (as litttle as one CM increments) while being capable of much more. You really think if you are vaulting ~20' feet in the air you can only possibly manage 1 cm more? Ha. He was playing the game.

He'd get the recognition, the accolades, and the cash sponsors payment over and over instead of just once if he just vaulted as high as he could. Nike alone paid him $40,000 every time he broke the World record. Intel is like Sergey. They have a lot more capability than they have been trickling out. They can pull a 3 year old part off the R&D shelf, blow the dust off, and have better performance than Threadripper, and undercut on price too - if they so choose.

Need more cores? I remember reading that Intel had processors in engineering with 64 cores back in 2011. At this point some of their production CPUs in the CPU space have 128 cores. Their best, certainty isn't something that gets bested by a 16 core newcomer. Intel's best just plays in very different markets. They need only rebrand some of their mid tier Xeon 20-30 core server processors as enthusiast processors and lower the price, unlock them and they are still years ahead of Threadripper + better IPC clock for clock.

I'm thankful AMD woke the Giant, but AMD is no Jack the Giant Killer.
 
Last edited:
horrorshow said:
It's called an "analogy".

School, much?
Click to expand...
Yes, an analogy, and a poor one at that. You still have not given anything to substantiate the common statement/belief some people tout that Intel has some secret tech they can just pull out and blow away the competition. It is the nature of the tech industry that this cannot be done. We would know about it years in advance, because that knowledge will be required to have the supporting ecosystem to go with the new product. The planned technology may turn out to be much more than expected (see Sandy Bridge CPUs) but the technology itself is no surprise.
 
Archaea said:
Take a lesson from Sergey Bubka.

They can pull a 3 year old part off the R&D shelf, blow the dust off, and have better performance than Threadripper, and undercut on price too - if they so choose.
Click to expand...
And how long would it take for that product to get to market and adopted? Certainly not quick enough, as there would need to be new chipset validation, new motherboards, etc., or massive re-engineering to work with existing products. I think people give Intel too much credit, when they have not ever shown they can grab something that isn't on their roadmap and release a competition killer. Definitely not an AMD killer, when most all of Intel's R&D has been to try to beat ARM instead.
 
jardows said:
And how long would it take for that product to get to market and adopted? Certainly not quick enough, as there would need to be new chipset validation, new motherboards, etc., or massive re-engineering to work with existing products. I think people give Intel too much credit, when they have not ever shown they can grab something that isn't on their roadmap and release a competition killer. Definitely not an AMD killer, when most all of Intel's R&D has been to try to beat ARM instead.
Click to expand...
They are already alive, and functioning in production markets -- just a vastly different segment. This isn't theory crafting. It's already been manufactured.

Here - google'd for you 128 core Intel --- look at those timestamps. AMD doesn't have that kind of product.
https://www.google.com/search?q=xeo...afe=active&q=intel+128+core&spf=1500583886663
 
You must log in or register to reply here.
Back
Top