N4CR
Supreme [H]ardness
- Joined
- Oct 17, 2011
- Messages
- 4,947
Zen+ already had higher IPC, might want to take that stock portfolio out of your eyeglasses.
![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
AMD possibly going to 4 threads per core
- Thread starter Ready4Dis
- Start date
misterbobby
2[H]4U
- Joined
- Mar 18, 2014
- Messages
- 3,814
Zen+ most certainly does not have higher IPC than something like the 9900k.
Saying that something isn't as important because hedt Intel had a version is a statement divorced from economical reality.
You just got to see the massive price difference between both teams solutions and heck even jumping on the same hedt space it's very obvious which one is way more bang for the buck both in upfront costs and when you calculate power consumption going forward.
Still this won't matter too much because manufacturing capability is more or less capped economy wise, meaning that there's no feasible way for amd to get to parity in volume of parts which is why even if Intel takes a couple years to iron his process wrinkles they should still be mostly safe.
Like our friend said this is a game changer for prosumers world wide tho even at current product volumes.
You just got to see the massive price difference between both teams solutions and heck even jumping on the same hedt space it's very obvious which one is way more bang for the buck both in upfront costs and when you calculate power consumption going forward.
Still this won't matter too much because manufacturing capability is more or less capped economy wise, meaning that there's no feasible way for amd to get to parity in volume of parts which is why even if Intel takes a couple years to iron his process wrinkles they should still be mostly safe.
Like our friend said this is a game changer for prosumers world wide tho even at current product volumes.
The difference is small enough to not matter. Intel just has the clocks. It is dumb to go anything Intel over AMD at this point. The new TR CPU are going to crush Intel. Sure for that small niche of low res high fps segment Intel still owns. Intel got nothing else
IdiotInCharge
NVIDIA SHILL
- Joined
- Jun 13, 2003
- Messages
- 14,675
You just went from talking about six-core and eight-core CPUs to twelve-core CPUs. Please, keep the crown you're failing to pass.
IdiotInCharge
NVIDIA SHILL
- Joined
- Jun 13, 2003
- Messages
- 14,675
I have plenty of stocks, none are of AMD nor Intel.
Zen+ had decent IPC -- Zen 2 has a little more. Neither eclypse Intel's aging Skylake arch.
In general this is true.
This point is dumb. There are plenty of reasons to go with Intel over AMD, at this point. Just not as many for enthusiasts.
Skylake-based Xeons? They had better. 10nm Xeons? Well, Intels IPC jump is already testing strong, so let's say that AMD is likely to remain competitive, which is high praise in that space.
Platform stability, maturity, software optimizations, oh, and there's the largest PC market, where AMD flat out has no product that competes: mobile. And yes high-end gaming. 9900K is where it's at if you're trying to maximize framerates and minimize frametimes and input lag. Paired with a 2080Ti, you're not going to get better -- this is also a space where AMD does not compete.
IdiotInCharge
NVIDIA SHILL
- Joined
- Jun 13, 2003
- Messages
- 14,675
Microsoft's scheduler for desktop Windows has been tuned for monolithic CPUs with equal 'cores'. The dual-die Pentium IV and Core 2 Quad variants have been the only exceptions in Intel's lineup, and Intel is the standard. AMD didn't do multi-die until Ryzen, and Bulldozer was fundamentally flawed in execution.
Basically, you're expecting Microsoft to account for whichever wacky thing AMD comes up with next and to be able to shoehorn existing applications into said wackiness. As evidenced by later benchmarks, they get there, but it takes time.
Also, while the Linux kernel was designed with significantly more flexibility in mind, and AMD can contribute code directly to it to ensure that the kernel is ready for their products, the story on Linux has been even worse with Ryzen.
This is on AMD.
IdiotInCharge
NVIDIA SHILL
- Joined
- Jun 13, 2003
- Messages
- 14,675
Started on Cyrix, actually- but that was back when AMD and Cyrix were copying Intel's x86 CPUs wholesale.
actually WAY before then even.
i had a NEC V20 that was 20% faster than the Intel 8088 it replaced
That is 100% untrue. Cyrix didn't have a licence until Intel gave them one to shut them up about Intels thieving ways.
AMD did for sure have some licensed clones early on (286 and earlier), but started deviating around 386 days.In fact Intel lost EVERY lawsuit they filed against Cyrix .... because no Cyrix chip ever was a copy in anyway at all.
Cryix counter suited and WON... cause Intel flat out ripped off Cyrix advancements when they designed the Pentium pro; you know the design that they have basically just been adding to for over 20 years now. Cryix ended that suit with a simple cross licence deal where Intel agreed to stop trying to sue Cyrix in return for Cyrix allowing Intels theft. Our absolutely shit legal system is responsible for the the fall of Cyrix. Cryix just wanted the legal bills to end... never mind that they where in the right every single time. They should have told Intel to spit and forced them to pay Cyrix royalties, in hind sight Intel would have owed Cyrix a royalty on every processor they have made since.
Cyrix contribution to CPUs has been reduced to platitudes in wikis. lol
From the Pentium pro Wiki page;
"The Pentium Pro incorporated a new microarchitecture, different from the Pentium's P5 microarchitecture. It has a decoupled, 14-stage superpipelined architecture which used an instruction pool. The Pentium Pro (P6) featured many advanced concepts not found in the Pentium, although it wasn't the first or only x86 processor to implement them (see NexGen Nx586 or Cyrix 6x86). The Pentium Pro pipeline had extra decode stages to dynamically translate IA-32 instructions into buffered micro-operation sequences which could then be analysed, reordered, and renamed in order to detect parallelizable operations that may be issued to more than one execution unit at once. The Pentium Pro thus featured out of order execution, including speculative execution via register renaming. It also had a wider 36-bit address bus (usable by PAE), allowing it to access up to 64 GB of memory."
No mention of Cyrix suing Intel over their theft and Intel settling. The 6x86 wasn't just first to those ideas Cyrix invented them. Intel is great at copying they copied their basically complete speculation out of order operation design from Cyrix and they where able to copy x86_64 thanks to previous court cases where AMD and Intel agreed to share. Intel surprisingly has very few original ideas... but with a massive war chest and 1000s of engineers they have been able to refine things and win the fab wars. (well till recently on the fab wars anyway)
Last edited:
That is 100% untrue. Cyrix didn't have a licence until Intel gave them one to shut them up about Intels thieving ways.
In fact Intel lost EVERY lawsuit they filed against Cyrix .... because no Cyrix chip ever was a copy in anyway at all.
Cryix counter suited and WON... cause Intel flat out ripped off Cyrix advancements when they designed the Pentium pro; you know the design that they have basically just been adding to for over 20 years now. Cryix ended that suit with a simple cross licence deal where Intel agreed to stop trying to sue Cyrix in return for Cyrix allowing Intels theft. Our absolutely shit legal system is responsible for the the fall of Cyrix. Cryix just wanted the legal bills to end... never mind that they where in the right every single time. They should have told Intel to spit and forced them to pay Cyrix royalties, in hind sight Intel would have owed Cyrix a royalty on every processor they have made since.
Cyrix contribution to CPUs has been reduced to platitudes in wikis. lol
From the Pentium pro Wiki page;
"The Pentium Pro incorporated a new microarchitecture, different from the Pentium's P5 microarchitecture. It has a decoupled, 14-stage superpipelined architecture which used an instruction pool. The Pentium Pro (P6) featured many advanced concepts not found in the Pentium, although it wasn't the first or only x86 processor to implement them (see NexGen Nx586 or Cyrix 6x86). The Pentium Pro pipeline had extra decode stages to dynamically translate IA-32 instructions into buffered micro-operation sequences which could then be analysed, reordered, and renamed in order to detect parallelizable operations that may be issued to more than one execution unit at once. The Pentium Pro thus featured out of order execution, including speculative execution via register renaming. It also had a wider 36-bit address bus (usable by PAE), allowing it to access up to 64 GB of memory."
No mention of Cyrix suing Intel over their theft and Intel settling. The 6x86 wasn't just first to those ideas Cyrix invented them. Intel is great at copying they copied their basically complete speculation out of order operation design from Cyrix and they where able to copy x86_64 thanks to previous court cases where AMD and Intel agreed to share. Intel surprisingly has very few original ideas... but with a massive war chest and 1000s of engineers they have been able to refine things and win the fab wars. (well till recently on the fab wars anyway)
100% all the way.
Cyrix reverse engineered Intel's chips and improved upon them, not needing to rely on licensing from Intel.
edit: PS - Too bad VIA destroyed Cyrix.
IdiotInCharge
NVIDIA SHILL
- Joined
- Jun 13, 2003
- Messages
- 14,675
I started on a Cyrix 486. That's true. It was a copy of Intel's 486. That's also true. That's what I stated, and it's 100% true.
I didn't say anything about licensing
I started on a Cyrix 486. That's true. It was a copy of Intel's 486. That's also true. That's what I stated, and it's 100% true.
I didn't say anything about licensing
It wasn't in anyway a copy other then it understood what x86 instructions where doing and converted them so that their RISC CPU could do the work.
x86 was a complete dead end. There is a reason Intel was going to kill it off in favor of Itanium. Cyrix showed the entire industry that an instruction set didn't have to be hard baked into silicon. x86 is upwards of 40,000 instructions that prior to cyrix where basically hardwired like an expensive scientific calculator. Their work proved you could build an intelligent prefetch code translator that would convert x86 into microop operations to be reordered and run more efficiently then simply executing the ancient x86 instructions directly. Intels own chips up to the Pentium simply decodes and performed x86 instructions. Cyrix did not build hardware to simply compute x86 instructions or they would have been sued out of business. They build a chip that converted those x86 calls into more efficient instructions its RISC core could compute... as a result they found ways to optimize the x86 instructions in hardware and execute them in parallel and out of order. It was a revolutionary design. Cyrix and NexGen where the two companies that pioneered that work. Intel stole it... and AMD bought speculative execution when they bought out NexGen for just under a billion in cash (in 90s money).
Anyway what I am saying is ya for consumers Cyrix 486s looked like a copy cause it could run the x86 instruction set... but the chip was actually a much more advanced design then Intels own 486. If the software industry had really gotten behind Cyrix they would have destroyed Intel in terms of performance. Especially when they got to 6x86 and had a full fledge speculation engine that was as good as anything Intel was using for 10 years after. However as of course you know from owning one that didn't happen... and Cyrixs chips where ya forced in translation mode 99.9% of the time.
Not completely related but as I brought up Nexgen... nexgen even had some proto chiplet style parts. I believe the process they used was an IBM one but a few of their later parts had math coprocessors in a separate die on package.
Makes you wonder how many interesting ideas could have been in use 20 years earlier if our court systems where different. Its just so easy for 1 or 2 large companies to keep the smaller upstarts burning their captial on BS cases.... just waiting for the day they can purchase them. Although looking back on that time we are probably all very lucky AMD managed to snag Nexgen before Intel did. AMD would no doubt be long gone if they hadn't made that purchase.
You might find this interesting IIC
https://en.wikipedia.org/wiki/Multi-chip_module
In case anyone thinks AMD was the first user of chiplets.
(not even the first x86 maker to use chiplets) Would be interesting to know how many AMD engineers are still around from NexGen I would assume not many... but perhaps a few are still around from those days. Of course Sue herself had some experience at IBM working with the IBM research dept which I'm sure had worked on some interesting chiplet stuff.
Last edited:
IdiotInCharge
NVIDIA SHILL
- Joined
- Jun 13, 2003
- Messages
- 14,675
"Let us compare the execution units of AMD's Ryzen with current Intel processors. AMD has four 128-bit units for floating point and vector operations. Two of these can do addition and two can do multiplication. Intel has two 256-bit units, both of which can do addition as well as multiplication. This means that floating point code with scalars or vectors of up to 128 bits will execute on the AMD processor at a maximum rate of four instructions per clock (two additions and two multiplications), while the Intel processor can do only two."
From my previous link, in normal (non high precision loads) AMD can run up to 4 simultaneous operations while Intel is limited to 2 in best case. In higher precision work loads they can both do 2 at a time. So (generalizing) AMD worst case is equivalent to Intel and best case is twice as good. In reality it appears to have a lead in smt usage. Obviously there is a lot more to this, it's very over simplified, but keeping the instruction pipeline/buffers full and data available is the hard part.
Red Falcon
[H]ard DCOTM December 2023
- Joined
- May 7, 2007
- Messages
- 12,452
Exactly, and the NEC V30 was ~18% faster than the Intel 8086, and both CPUs were easy drop-in upgrades.
I'm still running an NEC V30 to this day, and the performance boost over the 8086 was noticeable, especially in FTP transfers.
This was waaaay before both VIA and Cyrix, back in the early 1980s.
Yes it had the ability to translate 486 instruction calls into microops their RISC chip could compute. Had they simply etched 486 instructions into their CPU they would have lost their vs Intel law suit.
A good point. I am sure AMDs superior SMT performance is a combination of your point... their chips being able to better split less precise work, in combination with larger cache spaces that allow the same to hold longer prefetch histories to better split similar work.
OutOfPhase
Supreme [H]ardness
- Joined
- May 11, 2005
- Messages
- 6,579
Yes - I think we're agreeing. An increase in procunit efficiency only matters if it is consistently saturated. SMT is a tool to assist for an otherwise unbalanced design.
To clarify - I do not say "unbalanced" as a pejorative. It's a totally valid approach to have bulkier cores and then adjust the thread ratio to optimize efficiency. It's also valid to have a design with is more optimal closer to 1:1 T:C if you can more closely match the throughput of fetch+decode with compute/writeback.
As we've said in many such threads - isn't it nice to see competing effective designs again? Who wins? WE DO!
IdiotInCharge
NVIDIA SHILL
- Joined
- Jun 13, 2003
- Messages
- 14,675
...to the point that comparing raw compute units is like comparing engine cylinders in cars -- damn near meaningless without the rest of the powertrain specified, let alone the rest of the car.
And when you see that Intel still has IPC parity and even an IPC advantage in some workloads, with their outdated architecture, perhaps they made the tradeoff for a reason. Higher IPC, higher clockspeeds, and lower power usage at the low end (mobile), and they're essentially *two nodes behind, and about four years behind where they themselves planned to be. Again, I'm not knocking what AMD has accomplished with Zen, and I'm getting more hopeful that upping threads assignable per core will bring real advantages, but it's worth pointing out that this isn't a simple problem set with direct or easy solutions.
[*marketing nodes]
IdiotInCharge
NVIDIA SHILL
- Joined
- Jun 13, 2003
- Messages
- 14,675
And the AMD 486 I used?
Perhaps there is something to what you're saying -- however, it is also true that Cyrix and AMD were copying x86 designs from Intel early on. I did own an AMD 686 with 3DNow! later on, too. It was okay, but AMD sucked at FP until the Athlon, and then they decided to suck again with Bulldozer.
AMD had an actual licence. As I say ya those where basically clones... not only that they where built under contract from Intel to serve low cost markets. Their 386 was 100% a copy they where contracted by Intel to crate make them. Their 486 was their own design but again leaning heavy on their Intel licence. They paid royalties. (back then there was no guarantee that x86 was going to be the big boy forever... IBM power and MIPS ect could easily have replaced x86 if they got the right push, so Intel was happy to have low cost options around to head off MIPS especially) The last pure AMD cloner design was the K5... and although it had some "enhancements" and design choices some of which where of course to reduce production costs... they where still paying royalties.
Cyrix never had a licence, and never paid Intel one red cent. Their CPUs where NOT standard x86 chips. They where RISC chips that broke the x86 instruction calls down so they could be computed by their simpler RISC core. Turns out that was the future of x86 today neither Intel nor AMD execute pure x86 instructions in silicon as the 8086s ect did. They have a speculation engine that takes those calls and breaks them down into easier to compute chunks and are chunked by the RISC style cores at the heart of modern Intel and AMD chips.
AMD got a boost FPU and IPC wise after K6... which is when they bought out NexGen. NexGen like Cyrix was using a speculation engine to run x86 code on a RISC core. K6 was mostly a relabeled NexGen 5x86... later K6s got some AMD stuff included (3DNow) The K7 (Athlon) was the first time AMD took they had a chip Designed with both their own Engineers and the NexGen folks they acquired in their NexGen purchase (1.4 billion in 2019 dollars).
So ya back then Cyrix never ever copied an Intel design they built a RISC chip that could execute x86 commands. Its not just a legal difference... they literally invented modern x86. Around the same time Nexgen was also working on much the same tech and between Cyrix and Nexgen they basically had patented the 2 most practical ways to create a speculation engine. (There is a reason why L1 / L2 / L3 cache all sounds so much the same even if your talking about Power, ARM or RISC-V) Intel got a hold of the patents they needed to build the Pentium Pro by first just straight up stealing... and their lawyers made it ok by agreeing to basically leave Cyrix alone forever. (which was dumb on Cyrix part in hindsight... they should have pushed that suit to conclusion and they could have lived on forever with a nice juicy cut of every Intel processor sold for 20 years) AMD got into the same type of tech with a purchase. As I was saying we should all be thankful Intel wasn't smart enough to out bid AMD for Nexgen... had they done that Intel would have a monopoly on x86.
Red Falcon
[H]ard DCOTM December 2023
- Joined
- May 7, 2007
- Messages
- 12,452
IdiotInCharge
NVIDIA SHILL
- Joined
- Jun 13, 2003
- Messages
- 14,675
This is probably taking it too far. Yes, the idea of decoding x86 into micro-ops so that instruction-level parallelism and out-of-order execution may have predated the Pentium, but it was also really the only way forward. 'RISC' just means reducing the instructions and thus input complexity of the CPU ISA; everything is RISC these days.
Grimlaking
2[H]4U
- Joined
- May 9, 2006
- Messages
- 3,245
Man wow look at those old ass ISO cards.
If you remember Intel didn't believe that. Intel never built a RISC processor prior to the pentium pro... they where planning to jump form CISC x86 chips to EPIC (Explicitly Parallel Instruction Computing ). Itanium began development in 1989.
Intels plan was very much to go from 386/486/Pentium -> Itanium down the road when 64bit made sense in desktops. First in the server world and years later replacing their CISC Pentium chips. Even with the pentium pro they still went the whole P4 route. They didn't see the light until AMD forced the issue.
AMD threw a wrench in the works purchasing NexGen and then quickly getting K7 (Athlon) out the door. Intels plan was to run their long pipeline semi hybrid (CISC/RISC) P4 chips until desktops where ready for 64bit and Itanium.
I don't think I'm overstating things... Cyrix and NexGen made the first "modern" CPUs. No doubt Intel has refined the heck out of things 20 years later... if Intel had their way though we would have basically been using supped up 486s till the 2010s. lol
I agree, but the design decision has implications that give it the lead in some cases and equivalency in the others. I work with plenty of mcu's that do 1 instruction per cycle (ok, some complex instructions can take more), but those tend not to be the anywhere near fast enough for today's desktops. Overlapping instructions and deep pipelines are necessary to keep busy and be able to work on multiple opcodes simultaneously, but sometimes you have stalls due to waiting on results (aka, some things can't run in parallel). These are the instances where smt makes sense, or if your using one specific unit but not another, it allows multiple threads to run along side each other instead.of just stalling and leaving the pipeline under utilized. So it does depend on th cpu, decoder, memory system, cache and probably a ton of other things, but these things will happen in most workloads (just to a differing degree). As mentioned though, AMD has a more flexible arrangement and tends to lead in smt gains (in most workloads that can actually use that many threads, games aren't taxing 12 cores, forget 24 threads).
IdiotInCharge
NVIDIA SHILL
- Joined
- Jun 13, 2003
- Messages
- 14,675
I don't recall if the Pentium was RISC, but it did decode and do out-of-order execution. The Pentium Pro improved that a bit and brought it to a full 32-bit pipeline.
We'll probably go back to EPIC / VLIW. The idea pushes parallelism onto the compiler, which is why it was so difficult to get optimized code for the Itanium. Code that was optimized though, flew. And compiler design has come a long way; for AMD to be considering SMT4 for a consumer processor, they need to be pretty confident in software optimizations to make a difference, as otherwise their cores will stall on cache misses and context switches. Compiler optimization is the future of computing performance increases. Perhaps we should look for Linux kernel commits to see where AMD is going with SMT4?
K7 was equivalent to P6 with a tad more raw FPU, which while nice, wasn't a big leap. The leap was that AMD was competitive at all. Of course, they had supply problems and getting a K7 board for Slot A working was a shitshow. I don't know many that bothered. But Socket A, everyone jumped on that one. That's when Intel took a left turn and instead of giving us Tualaten CPUs (updated Pentium IIIs), they went with the RAMBUS-equipped Willamette. You were still lucky to have a stable AMD system, but you could afford it, and it was competitive.
Notably, K7 and its successors didn't improve in terms of IPC nearly as much as Intel improved the P6 once they decided to shift back in earnest. Literally the first new P6 had people clamoring for more, and once Core 2 hit, AMD was done until Zen+. They still haven't pulled out of second place IPC.
If Cyrix's CPUs had been that good, they'd have taken the market. They weren't. I ran them. They worked, but they were weak. Same for AMD up until K7 -- I used its predecessors. Also weak. Ran plenty of K7s too though.
This I don't get. Intel has been innovating like crazy, and hasn't stopped. AMD bought the team that brought them the K7; they refined it a bit, brought the memory controller onboard, expanded the registers to 64bit at a strategic point in time, shrunk it and pumped up the clocks a little, but ultimately abandoned it for something... worse. They had their own Netburst moment, right after they kicked its ass. Excuse me if I don't put much faith in them. I want them to succeed, but I also spend so much time just getting their damn boards to function too -- and Ryzen has been no different. Ran tons of ATi GPUs, cried a little inside when AMD bought them, ran a few -- and spent plenty of time with several AMD GPU fuckups. And I still buy them and recommend them.
But I'm under no illusion just how AMDs brilliance and release shitshows go hand in hand. I don't expect SMT4 to be any different, even as I understand that there's plenty of room there for them to carve out more of a performance niche.
And while they do that, remember that Intel has been putting out their top-end desktop CPUs in the US$300 - US$350 range for over a decade now. They've increased clockspeeds, increased IPC, increased the performance and stability of their GPUs and expanded the capabilities of their GPUs beyond what even AMD has done -- and they've led in Linux kernel development. Intel has worked both to increase performance and decrease price / performance even as AMD took a left turn into irrelevance for a decade. Literally competing with themselves, and yet still serving customers.
AMD still doesn't have an answer for laptops -- they're two or three generations behind Intel in mobile -- and yet that's where Intel has put their newest technology, not just new CPUs, but also new graphics, chipset, and wireless capabilities. And I'm over here wishing for a full-blooded AMD APU and accompanying featureset for my next laptop upgrade. Unfortunately, I've watched AMD long enough to know not to be too optimistic.
Rockenrooster
Gawd
- Joined
- Apr 11, 2017
- Messages
- 955
Overall, Zen 2 has higher IPC...
The only scenario where Skylake can have higher IPC is in gaming workloads. In everything else Zen 2 has higher IPC, by a measurable margin.
Intel's advantage is having clocks, thus it has slightly higher per-core performance in some workloads, but not all.
Your comment only rings true if the only measurement you go by is gaming type workloads. Or avx512 I guess...
Rockenrooster
Gawd
- Joined
- Apr 11, 2017
- Messages
- 955
I just find saying that statement silly even though you know better.
People that pay attention to these kinds of things (process nodes) would know that Intel's 10nm is comparable to TSMC's 7nm, and thus Intel would only be 1 node behind in effect.
Process woes or not, it's Intel's fault they are behind, they own their foundries and have crap tons of money, there is almost no excuse to be in their situation right now except for pure laziness and collecting profits. Sure they were making tons of money, but what the heck were they doing with all that money??? Sure I know that they do more than CPUs and IGPUs (AI crap, etc), but with the money they were making, they should be able to do so much more. They're just now trying to trim the fat and become better.
It boggles my mind how AMD can compete with how little money they make in comparison in not only CPUs, but in GPUs as well...
Last edited:
" Unfortunately, I've watched AMD long enough to know not to be too optimistic"
Too bad, guess you missed Zen, Zen2, Epyc, Threadripper, I/O chiplet development, Infinity Fabric, x64, first 1GHz proc.
What's Intel doing?
Oh yeah,
-> brought the old Keller guy from where?
-> brought that throw away Raja from where to fix their GPU division?
Too bad, guess you missed Zen, Zen2, Epyc, Threadripper, I/O chiplet development, Infinity Fabric, x64, first 1GHz proc.
What's Intel doing?
Oh yeah,
-> brought the old Keller guy from where?
-> brought that throw away Raja from where to fix their GPU division?
Rockenrooster
Gawd
- Joined
- Apr 11, 2017
- Messages
- 955
To be fair, Jim Keller does whatever the eff he wants on his terms, he bounces back and forth from a lot of high profile tech companies and has worked for some more than once...
True, but the point is that Keller made his name at AMD twice beforehand - K7/K8 and then came back to AMD for Ryzen.
But this confuses me, AMD had a licence for CISC silicon, Cytrix didn't, but didn't need it... Then why do people keep saying AMDs x86 license is not transferrable or some such? If AMD is really RISC silicon with translation, then anyone can do that? Anyone can go to AMD, licence their shit, and have an AMD junior CPU that is x86, just not CISC... I mean what is an x86 license anyway?
That has me a little confused.
AMD has a cross-license with Intel. Complicated nowadays with AMD's x64 cross-licensed by Intel and parts of Intel's x86 cross-licensed by AMD. Correct - it is NON-transferable. Just as Intel's x64 license to AMD will not transfer. An incestuous mess.
Cyrix just did an in-house reverse engineering of the Intel chips and then enhanced/improved the design - therefore did not need licensing.
->Microcode is copywrited.
Hopefully soon, looking at 2020 though, hopefully first quarter, but that's just my optimism speaking
.https://www.google.com/amp/s/amp.to...-renoir-apu-zen-2-vega-10-graphics,40145.html
https://amp.hothardware.com/news/amd-ceo-confirms-radeon-navi-zen-2-mobile-cpus-incoming
https://en.wikipedia.org/wiki/P5_(microarchitecture)
" The P5 Pentium was the first superscalar x86 processor; the Nx586, P6 Pentium Pro and AMD K5 were among the first designs which decode x86-instructions asynchronously into dynamic microcode-like micro-op sequences prior to actual execution on a superscalar microarchitecture; this opened up for dynamic scheduling of buffered partial instructions and enabled more parallelism to be extracted compared to the more rigid methods used in the simpler P5 Pentium; it also simplified speculative execution and allowed higher clock frequencies compared to designs such as the advanced Cyrix 6x86. "
The pentium was super scalier but it was still a CISC processor.
We can leave it there 20 years ago Intel ripped off Cyrix. Cyrix was no clone, Intel has/had plenty of smart engineers they recognized a revolutionary design when they saw it. They copied a ton from Cyrix stepped on their patents... and got through it cause they had Cyrix in court wasting their money for almost 10 years already at that point. Cyrix just wanted to legal BS to end so they let Intel get away with it in return for Intel swearing off the lawyers.
I am not saying home users where buying superior chips with Cyrix... just better designed chips. Cyrix still had multiple things working against them... they used chipsets designed for another companies part, they where pretty much always fabbed on less then processes compared to their Intel counterparts, and to really unlock some of their superior design software would need to be optimized for Cyrix. Of all those issues where only solvable with lots and lots of cash... something they didn't have, and one big reason was Intel suing them over and over. Even though Intel lost every single time it was still very expensive.
Again I'm not denying that for consumers Intel was the superior product for a long time... just that Intel as has been the case until pretty recently has leaned on their massive $ war chest to maintain a fabrication and marketing lead. When there fabs where ahead they made hay... when they where not they paid off OEMs, and reminded them that in the future they would be ahead in the fab race again so they better keep Intel happy.
Anyway to bring it back to the thread.... SMT4 or 8 or 16 doesn't require any more optimization then you would expect as long as its implemented well. The power chips with SMT8 actually require very little in terms of optimization compared to standard power code... the catch is IBM designed them with a ton of cache. SMT4 on a Ryzen chip with 64MB of L3 cache on a 8 core part may well perform a lot better then most of us would expect. Still think this rumor is 50/50 at best of being true... just saying I do believe AMD could pull off something interesting based on what we have seen from the last few generations of Ryzen.
https://en.wikipedia.org/wiki/AMD–Chinese_joint_venture
Well ya AMD can licence their x86 stuff.
AMD had a licence they got in the game after being contracted by Intel to produce low cost x86 chips. Intel was happy to have another company making cheaper x86 chips. Early on Intel was not manifest CPU destiny... there where TI chips IBM chips MIPS chips, and those where just the big 3 other players. There was a time when many thought Unix would become the major home desktop OS powered by all manner of different TI/MIPS/Power and even x86 chips. Later as things became more Win/Tell it made goverments look at them as less then a monopoly cause there was another company making x86. Having said all that Intel did get a little annoyed with AMD around K5 or so as they felt AMD was starting to cannibalize more mid range market then they liked... and they where driving the price on their higher end stuff down. Intel tried to fix this by producing crappier low end chipsets (which backfired as VIA and others got into chipsets cutting Intel out completely)... and it was around this time that AMD realizing K5 was not so great and they where lacking some tech bought Nexgen and not long after released K6. This move caused Intel to flip their lid and not long after they moved to kill the idea of sharing sockets and we got slot 1 Pentium 2s. From that point on AMD became less and less a clone company and started doing their own thing. Which is why we all remember our Athlons so fondly. It was the first time a non-intel chip was noticeably better then Intels chips and not just cheaper.
Cyrix didn't have a licence. Intel didn't hire them to make cheapo x86. Cyrix white roomed x86. I'll just post the crib notes from wikipedia;
" Unlike AMD, Cyrix had never manufactured or sold Intel designs under a negotiated license. Cyrix's designs were the result of meticulous in-house reverse engineering and often made significant advances in the technology while still being socket compatible with Intel's products. In Cyrix's first product, the 8087 math co-processor, Cyrix used hardware math multipliers rather than the CORDIC algorithm, which allowed the chip to be faster and more accurate than Intel's co-processor. Thus, while AMD's 386s and even 486s had some Intel-written microcode software, Cyrix's designs were completely independent. Focused on removing potential competitors, Intel spent many years in legal battles with Cyrix, consuming Cyrix financial resources, claiming that the Cyrix 486 violated Intel's patents, when in reality the design was proven independent. {Rulings from federal court in Sherman, Texas and Federal Circuit Court of Appeals in Washington DC.}
Intel lost the Cyrix case, which included multiple lawsuits in both federal and state courts in Texas. Some of the matters were settled out of court and some of the matters were settled by the Court. In the end after all appeals, the courts ruled that Cyrix had the right to produce their own x86 designs in any foundry that held an Intel license. Cyrix was found to never have infringed any patent held by Intel. Intel feared having to face the antitrust claims made by Cyrix, so Intel paid Cyrix $12 million to settle the antitrust claims right before a federal jury in Sherman, Texas was to hear and rule on the antitrust claims. As a part of the settlement of the antitrust claims against Intel, Cyrix also received a license to some of the patents that Intel had asserted that Cyrix infringed. Cyrix was free to have their products manufactured by any manufacturer that had a cross license with Intel, which included SGS Thomson, IBM and others. {Rulings from federal court in Sherman, Texas and Federal Circuit Court of Appeals in Washington DC.}
The follow-on 1997 Cyrix-Intel litigation was the reverse: instead of Intel claiming that Cyrix 486 chips violated their patents, now Cyrix claimed that Intel's Pentium Pro and Pentium II violated Cyrix patents—in particular, power management and register renaming techniques. The case was expected to drag on for years but was settled quite promptly, by another mutual cross-license agreement. Intel and Cyrix now had full and free access to each other's patents. The settlement didn't say whether the Pentium Pro violated Cyrix patents or not; it simply allows Intel to carry on making products under a license from Cyrix. "
As far as I know Intel is still using the patents it acquired from Cyrix. As to why others haven't tried... they have. Transmeta was another example. At this point the bottom line is... x86 is only really any value for desktops, and that market is shrinking and you are taking the chance of being sued by Intel or AMD at this point if you step on any patents that involve x86 silicon translation. Never mind breaking into the market where even AMD has a hard time breaking in with OEMs and they have been around for 40 years. The fact is everyone has gotten into CPUs... they are just going the much much easier ARM round. Licence ARM build what you like... Apple, Samsung, TI and many others are producing ARM. And no matter what we think of x86 the fact is the vast majority of compute devices in the world today are ARM not x86. There is even a lot of RISC-V work happening... even if most of that will always just be co-processor type stuff such as what Western Digital is doing with RISC-V.
Last edited: