Nearing the end of the cycle for their current generation products, its not surprising to see poor financial results come out from AMD. Last year was a terrible one and it looks like 2016 won’t be much better, at least for Q1. For the first quarter of 2016, AMD posted a net loss of $109 million from an operating revenue of $832 million. Unsurprisingly, it is better than 2015 as that year was arguably the worst ever.
AMD blames the revenue drop of 13% sequentially and 19% year-over-year as lower semi-custom sales. This is somewhat expected as we continue the PS4 and Xbox One lifecycle. The bright side is that Sony is set to release the PlayStation 4 Neo and even the Xbox One will see new revisions if not a full upgrade. Combined with the Nintendo NX, those should bounce back the semi-custom segment as consumers buy more consoles again.
Even though margins improved slightly to 32% (Intel posts around 60%), the increase in expenses led to the loss. This is reportedly due to increased R&D for upcoming products, which in my mind are due to Vega/Navi and Zen+ since Zen and Polaris are all set in stone by now. With Polaris 10 and Zen coming this year and even an Apple deal in the works, AMD has a good chance to turn things around as long as they can execute and head back to the black.
Back at E3 2015 nearly a year ago, AMD showed off their Project Quantum PC featuring 2 Fiji GPUs in a tiny form factor. Ironically, the feature AMD device used an Intel CPU instead of an AMD one and ended up using a single Fury chip instead of the dual Fiji we have come to know as the Radeon Pro Duo. Along with supply issues, we likely won’t see Project Quantum for a while. According to Diit though, when it does arrive, it will use AMD’s own Zen CPU and new Vega GPUs.
The main reason AMD chose to use an Intel CPU was simple. AMD CPUs were not up to snuff and with the Project Quantum aimed at being the best, it required a top-end CPU, one from Intel. With Zen set to debut later this year though, AMD has a chance to showcase the potential of their chip, showing that is capable of driving the fastest graphics cards out there without holding anything back.
On the graphics side, the delay on the CPU side means Vega, the full-on Fiji replacement with HBM2 will have a chance at Project Quantum. Vega should have no trouble beating FuryX and potentially even the Radeon Pro Duo. By delaying, AMD also reaps the benefits of moving the entire system to 14nm FinFETs, finally making the true VR PC for those that want the best.
UPDATE: We have now been informed from a kind reader that this was an April Fools joke, so please disregard the benchmark images below
AMD has been lingering behind in the enthusiast CPU market and really struggled to compete with Intel’s flagship products. This isn’t a shocking revelation when you consider AMD is still using the FM2+ socket to house its current processor line-up. Thankfully, Zen is upon us and the first major socket change in a considerable amount of time. We’re all hoping that AMD can become competitive again and Zen really helps bring innovation forward in the stagnant CPU market. AMD’s President and CEO, Lisa Su provided a small insight into Zen’s performance numbers and suggested they will bring a 40 percent IPC boost over the current line-up. Up to this point, all performance benchmarks have been kept under wraps and any numbers revolved around pure speculation.
However, images provided by Bits&Chips clearly illustrate the performance differences between a Octacore AMD Zen CPU and competing products. The CPU’s FP32 Ray-Trace score outperforms the i7-4930K but it’s not as impressive as the CPU hash results. This means the architecture might implement a weaker FMA.
On a more positive note, DDR4 bandwidth performance is impressive and competes with the i7-6700K. The CPU Hash is significantly better than the i7-5820K and even surpasses a 20-core Xeon. Only time will tell if AMD’s latest processors can offer similar performance to Intel products and instigate a pricing war. Currently, the i7-6700K is extremely expensive for a 4-core CPU and there needs to be some competition to drive innovation. I cannot wait to get my hands on AMD’s AM4 motherboards and finally see if they’ve come up with the goods. The basic data we have so far and information from AMD is promising but it’s always unclear until the testing has been completed from independent sources.
Do you think AMD will be able to have a much stronger foothold in the CPU market once Zen releases?
AMD’s upcoming Zen architecture is arguably the most anticipated hardware release this year. After years in the wilderness, AMD will finally come back with a new CPU design that will challenge Intel again on IPC, process node and power efficiency. According to the latest leak, it appears that Zen is progressing well enough that engineering samples have already been distributed to various partners for testing. This also means AM4 motherboards are already sampling as well.
These stepping A0 samples are that of the previously rumoured 95W, 8 core Zen CPU. That AMD has managed to get an 8 core CPU in a 95W thermal envelope is stunning and combined with the early engineering sample release, points to a strong 14nm LPP process. What’s more, the frequency isn’t a slouch, at 3Ghz base though boost isn’t enabled yet. This is pretty much the same as the base clocks for Intel’s own prosumer i7 5960X which sports 8 cores as well at 3Ghz base and 3.5Ghz boost. We can expect the Es to set the baseline so release Zen will almost certainly clock higher.
At 3GHz, the engineering sample is already faster than the first Bulldozer ones suggesting that 14nm LPP won’t be holding back frequency too much. After all, Intel’s own 14nm process has performed better than their 22nm. Samsung and Global Foundries have also had plenty of time to refine their 14nm process to ensure it will offer the best performance at launch. Hopefully, AMD will be able to be competitive in both IPC and overclocking.
With each passing week, it seems like more rumors are coming concerning AMD’s upcoming Zen processors. From previous reports, we had expected AMD to release Zen near the tail end of 2016 but revenue from the new CPUs was not expected till early 2017. According to the latest rumour though, AMD may have Zen out a bit earlier than expected, with an 8-core chip coming out as early as October, meaning there could be significant revenue from Black Friday and the holiday season.
An October launch pretty much falls smack dab in the middle of AMD’s late 2016/Q4 confirmed launch window. With an 8-core chip with potential SMT (Simultaneous Multi-Threading) for 16 threads, AMD is striking with a strong with a very competitive chip. Currently, Intel only has one consumer 8-core in the 5960X which is pretty pricey with a large TDP. An 8-core variant would also entice AMD users to consider an upgrade earlier as it doubles the thread count of current Piledriver CPUs on top of the 40%+ IPCincrease.
Launching under the Summit Ridge series, the 8-core also boasts a positively regular TDP at 95W. This considers favorably with what Intel has to offer, especially considering AMD is offering double the core count. This is probably due to toned down clock speeds, improved efficiency and the new 14nmLPP process, providing AMD with a jump of 3 process nodes. Intel, for instance, cut power by about 30%+ when they moved from 32nm to 14nm, at the while increasing performance.
Key to Zen, of course, will be how well it will perform. Based on AMD’s figures and what we have been able to glean from multiple leaks, Zen should hover around Haswell levels of IPC. The bigger question is how AMD prices Zen, though many will undoubtedly jump at the chance to buy a once again, IPC competitive AMD CPU. Hopefully, AMD’s expectations for Zen hold true. I for one, am finally hoping we will see 8C/16T CPUs enter the realm of mainstream hardware.
First off, Zen will introduce a new L0 cache, meaning that there will actually be 4 levels of cache. The L0 cache is a uOp cache, something Intel added back with Sandy Bridge. Paired with the uOp buffer, this will help reduce power consumption when running loops or if something needs to be re-executed quickly. Intel’s cache is 1.5KB so we can probably expect AMD to follow similarly as speed is more important than size.
Next up are changes to the L1 Instruction and Data caches. The L1 I$ will be 32KB, a drop compared to Steamroller/Excavator and K10 but back to the same size as Piledriver. The L1 D$ is also expected to be 32KB, a doubling over Steamroller and the same that of Excavator though still lower than K10. The reduced L1 I$ may be offset by the new uOp cache. The L2 may remain the same since the days of K10, with 512KB. This may be a problem if the rumoured inclusive cache design is used as 2304of the rumoured 8MB of L3 will be used in duplicating data. Having everything duplicated in L3 may make for better core-sharing and multi-threaded performance but limits everything to near L3 speeds for cache writes.
Overall, the cache changes suggest a move to ensure faster, rather than large caches. The increases to the caches also point to the focus on keeping the cores fed as well as high-speed cores with a long pipeline. This all helps with the 40%+ IPC improvement AMD is hoping for with Zen. Overall, Zen is looking to be a very wide and balanced design, borrowing from Intel and K10 but without any of the baggage of the past.
For those are hoping for “MOAR COARS”, it looks like AMD will be delivering later this year. First alluded to in a Linux patch last week, AMD’s upcoming Zen Opteron CPUs are set to have up to 32 psychical cores. A leaked slide from CERN reveals that patch was right on target. Combined with the introduction of Symmetrical Multi-threading, this will allow Zen to handle at least 64 threads at once, an unprecedented amount for AMD and quadruple current chips.
In addition to the large core, Zen is expected to bring PCIe 3.0 and DDR4 to AMD’s server offerings. The memory subsystem also gets a major boost with up to 8 channels, double from the current 4 on Socket G34. Compared to Intel’s Haswell-EP, Zen will offer 14 more cores and 28 more threads and double the memory channels. While Broadwell-EP may change things up later this year, AMD may still hold a lead in terms of core and thread count.
Combined with the expected 40% IPC boost, Zen may finally bring AMD back into relevance in the lucrative server and data centre market. AMD has had no real update to their server lineup since 2011, leading to their market share dropping to near zero. With such a major update, AMD will once again be competing in the server market with Opterons that can go toe to toe with Intel. While 32 cores is unlikely for the consumer lineup, a 16 core chip seems pretty likely.
With Zen set to arrive within the year, AMD is starting to get software ready for their revolutionary CPU architecture. According to a new Linux patch, AMD not only has Zen planned for this year but also a previously rumoured but never confirmed Zepplin CPU. Most interestingly is the connection between Zepplin and Zen which takes 3 letters from Zepplin, suggesting that they may be released, just as Piledriver and Vishera were also related for AMD.
According to the patch, Zepplin is part of the AMD’s Family 17h series of chips. This follows the previous 15h Bulldozer and its derivatives and the stellar K8(8h) and K10(10h). The 17h family is also expected to be the one Zen belongs too, with the K12 being the ARM-based lineup coming soon as well.
On Zen, the patch reveals some new information about how the LLC or Last Level Cache, usually the L3. Each Core Complex or Core Cluster of 8 threads shares 1 LLC unit. Given the use of SMT, we can expect Zen to be based on clusters of 4 cores. This is quite expected as larger CPUs use interconnects to link various different clusters of cores and their LLC together in order to function as a larger core. While Intel’s implementation carries a low penalty for accessing the LLC of another cluster in the chip, AMD’s L3 has been less robust so Zen will hopefully remedy this. You can see Intel’s Xeon E7 v3 design below with clusters formed by 2 cores and linked through buffered switches and interconnects.
Finally, Zen is also expected to feature up to 32 physical cores possible with 64 threads in total. This suggests that the massive Zen CPUs rumoured may actually come to fruition. This should allow AMD to take back some of that lucrative data centre and supercomputing market. All that remains is for Zen to finally deliver and bring AMD back into the black.
With their 14nm LPE (Low Power Early) process behind them, Samsung has moved on and started mass production for their new 14nm LPP (Low Power Performance) process. This is very important for the PC market as AMD is slated to use the 14nm LPP process from Samsung/GlobalFoundries for their upcoming Zen CPUs and Polaris GPUs. A ramp up now means that any major issues with 14nm LPP have been fixed and chips can start being produced.
Unlike the earlier LPE process, LPP is meant for higher power parts that clock faster, perfect for use with CPUs and GPUs. While AMD has not confirmed if they will be using Samsung or GlobalFoundries, this time around, the tight partnership between the two fabs means that a successful ramp for one is a success for the other partner. So even if Samsung is focused on producing their own mobile chips, GlobalFoundries is probably ready for Zen and Polaris.
Once Zen and Polaris production starts up, it will take around six to eight months for products to become available to us. This fits right into the timeline for when Zen and Polaris will launch in 2H 2016. Keep in mind that since Polaris will also use TSMC’s 16nm process, Polaris could arrive before Zen. With 14nm LPP on-track and Polaris and Zen taped out, it is only a matter of time before we see AMD launch their new lineup.
After releasing their poor 2015 financial results, AMD does have some more good news coming out for 2016. As most of you will already know, AMD’s upcoming Zen CPUs will launch and become available sometime in late 2016, likely towards the tail end. As part of the financial release, CEO Lisa Su revealed some key details about the Zen timeline. Chief among them is that Zen will be available for OEM partners to ship high-end PCs before the end of the year.
For the past few launches, AMD has suffered from issues of short initial supply, driving up costs for fans and other early adopters. This meant even if AMD has a great product, customers simply could get it and others may be turned away by the inflated prices. OEM partners tend to require and ship a whole lot more CPUs and systems than DIY builders. If AMD is able to supply the OEMs with adequate numbers of Zen chips, it should mean that DIY builders will also be able to purchase them without issue, making for a smooth launch with plenty of product.
OEMs being able to ship Zen systems by years end also points to the fact that Zen will likely be well on time and any issues with the new 14nm LPP process have been resolved. With over 40%+ IPC increase, Zen may finally be what’s needed for AMD to make a comeback in the CPU market.
AMD just can’t seem to catch a break these days. After seeing Nvidia surging ahead in GPU shipments for the past year and Intel continuing its dominance, we are now seeing it effects on AMD’s 2015 financials. Compared to 2014, 2015 was dismal, with a 28% drop in revenue from $5.5 billion to $3.99 billion this year. More importantly, gross margins fell a few percentage points leading to $419 million loss for the year.
2015 stands out in particular as AMD had managed to earn a few million in 2013 and even earn a hundred or so million in 2014, one of their past years in a while. Those successes were driven by the console royalties from Sony and Microsoft and strong GPU sales. This past year has seen console sales continue to slow as their lifecycle grows older, while AMD has released many new GPUs of their own. The GPU side had long been the major income earner and Nvidia hit that hard with their popular GTX 970 and 980 cards with the 970, in particular, being exceptionally well selling. AMD did note that they gained back some market share late in the year with their new Fury lineup.
Lack of new CPUs and GPUS led to a 42% drop in Computing and Graphics revenue, largely accounting for the year’s loss. 2016 should prove to be much better as expenses are expected to stay the same while new products like Zen and Polaris should drive up revenue. Polaris should help bring AMD back into balance with Nvidia while Zen will finally revive the ailing CPU side. With results as bad as this, it can only get better given what’s coming down the pipeline.
After many years of lagging CPU performance, AMD has suffered diminishing market and financial performance. This may change later this year as AMD comes back with their new Zen architecture that AMD originally expected to bring up to 40% IPC increase over their current lineup. According to AMD’s latest report, though, Zen may be performing better than their early estimates, with greater than 40% IPC increase over the current generation.
AMD has long been confident that Zen would deliver on its performance gains. After all, the new architecture is reported to bring instruction set parity with Intel and a reworked and more balanced design. In fact, rumours about a Zen Apple chip have surfaced, which if true, is a strong endorsement for AMD. Zen will also revamp the ageing 9xx series chipset and bring DDR4 and other new technologies with AM4.
While there is much to be optimistic, the biggest disappointment is that the rumour about an early 2016 release is wrong, with the originally suggested late 2016 launch being confirmed. This is despite the tape out having already been completed. The wording for the 40%+ IPC increase also suggests that the increase won’t be much greater than expected, otherwise, numbers like 45% or 50% would have been used. Hopefully, Zen will have what it takes to bring AMD back to the forefront and allow the company to continue with future releases and even stunning designs like this one.
In the days before AMD launched their APUs, all of their consumer CPUs largely used the socket across their lineup. When AMD launched their Llano series of APUs in 2011, they used the an incompatible FM1 socket instead of the AM3/3+ due to the need to integrate the iGPU. With the subsequent Trinity and Richland APUs, AMD kept a different socket in FM2. This year AMD is finally moving to a unified socket.
Officially confirmed as AM4, this new socket will combine the CPU and APU lineups for AMD. This means users will no longer have to decide which platform as well as chip they want, simplifying the decision to between a CPU or APU. This means users can purchase an APU on a budget and upgrade to a dedicated CPU and GPU later on without having to buy a new motherboard. This should help drive sales of AMD chips since it simplifies choice and offers more flexibility.
In bringing the two platforms together, we can finally expect to see Zen CPUs become a SoC. This is because the APU lineup already has the PCIe lanes tied to the CPU directly which bring along power savings and better performance, something the CPU lineup will finally pick up. Zen is expected to bring DDR4 support along as well. The biggest questions will be AMD have another socket meant for higher end chips like Intel does, whether or not the Zen CPUs will have some form of on-die graphics like Intel does and if AM4 will still be PGA.
Ever since Llano launched back in 2011, AMD has been pushing their APUs as being the next big thing. Combining a powerful CPU and GPU on one die, the APU allows even budget users to enjoy strong graphics to play games or use it for compute. In 2016, AMD is looking to combine their new Zen CPU architecture with their Polaris based graphics and it looks like something revolutionary will happen. According to a leaked paper, AMD may be planning a multi-core Zen APU coupled with a massive iGPU and an HBM cache with 128GB/s of bandwidth
For AMD, APUs have stayed strong even as their CPU line faltered, with their iGPU beating out Intel solutions. This changed though with Intel’s Iris Pro graphics that came with an eDRAM cache, with the high-speed cache helping put Intel neck and neck with AMD’s top APUs. The only bright side for AMD is that Intel’s chips have cost significantly more than AMD’s offerings. With the launch of a competitive CPU architecture in Zen, a new GPU architecture in Polaris and a large HBM cache, AMD has a chance to surge past Intel in this critical segment.
At a peak of 128GB/s, the HBM cache performs similarly to that of the memory of graphics cards like the GTX 760 and 960. If AMD adds in a high-performance iGPU, the APU has the chance to match or even exceed $150 graphics cards. With this APU, AMD is truly bringing gaming to the masses. Maybe in 2016, we may finally see AMD APUs powering Apple products.
AMD’s new line of CPUs and APUs are hotly awaited amongst hardware enthusiasts and long-time fans of the company. These new chips, set to make use of AMD’s upcoming AM4 socket have been the subject of a number of leaks, the latest of which saw a number of AM4 products being spotted on a leak from Zauba’s shipping database. The leak included mention of multiple components, with both an AM4 quad-core CPU, quad-core GPU and also an AM4 motherboard.
The leak shows that the components had been shipped to AMD’s testing facility in India, and WCCFtech believe that they belong to the Bristol Ridge family of APUs. As can be seen, 3 different batches of deliveries were made in late October and early November. Included in these were a batch of AM4 APU prototypes and a further batch of AM4 CPU prototypes, both with a TDP of 65W. The later of the shippings included an AM4 motherboard with the FOC acronym, MYRTLE codename and mentioned DDR4 support. FOC stands for Full Operational Capacity, meaning it is likely that the AM4 motherboard specifications are complete.
When you start to aggregate the data on AMD’s leaks, it starts to paint a picture of just when the chips could be released. A leaked BIOS microcode update three months ago showed the addition of functionality for AMD’s new AM4 chips. And WCCFTech notes that the last leak on a chip sent for testing, the mobile chip Carrizo, saw the chips being released just 4 months later.
This all could mean that AMD’s new shot at the gaming CPU crown could be on the way soon, both with its high-end Summit Ridge chips, and more budget all-in-one Bristol Ridge and Stoney Ridge APU chips. The question of whether AMD will be able to oust Intel from their position at the top, which has seemed secure for many years now, and return to the days of the Athlon remains to be seen.
With Zen launching sometime next year, AMD is casting their eyes towards its successors, Zen+ and beyond. While details on either have been scarce, AMD is hoping it that the new architecture will be springboard back into the much more lucrative data centre market. AMD peaked in that segment at 25% marketshare, but that has eroded away to less than 1% now.
According to CEO Lisa Su, Zen is going to start a 3 to 5-year push into data centers. Using the new hardware, AMD is building up a family of cores to ensure the platform as longevity. For the consumer side of things, this probably means that the new AM4 platform will be here to stay for a while. AMD has been known for keeping sockets around for longer periods of time than intel and with the new AM4 platform will likely be very forward-looking.
If AMD is already working on successors, it hopefully means that Zen will be launching very soon. It also hopefully means that AMD is confident that it will deliver enough and be able to evolve in the long-term. One of the biggest issues with Bulldozer and K10 were that those designs couldn’t evolve. Hopefully, the new chips will form a good foundation for AMD to evolve off of, just like how Pentium 3 worked for Intel.
Documents have surfaced, via Benchlife.info, that suggest AMD is starting its transition from Excavator architecture to Zen architecture, with the company’s new Socket AM4 arriving on new motherboards by March 2016.
AMD has been using its Socket AM3 for over six years, so is well overdue an upgrade. The AM4 socket will support both Zen CPUs and Bristol Ridge APUs, plus DDR4 RAM and future FX CPU and APU support. DDR3 will not be supported, however. The 14nm Zen processors will support Simultaneous Multi-Threading Support Technology (SMT), allowing a performance increase of up to 40% Instruction Per Clock (IPC).
Recent unverified reports suggest that the Zen architecture has been fully tested by AMD and has “met all expectation[s]” with no “significant bottlenecks”, with hopes high that the new processor line could rejuvenate the ailing chipmaker and be more “competitive against Intel” following the relative failure of its Fury GPU series this year.
AMD’s Zen architecture, built on the company’s new 14nm process, will prioritise increasing per-core performance over core count and multi-threading, and will sport 95W TDP.
Before anyone gets too excited about this news, so far this is only a rumour that originates from one source, but that doesn’t mean that we can’t dream, hope and get a little bit excited. A german tech publication got wind from a Motherboard manufacturer that we could see new AMD Zen based motherboards as early as March 2016, well ahead of the otherwise known Q4 2016 plan.
The new AMD Zen CPU will use a new socket and also incorporate the current DDR4 capabilities and as such completely new boards have to be designed. In a routine inquiry about this, the publication got a surprising response from the manufacturer. The response: “The internal timetable foresees a March 2016 launch”. That is a lot sooner than expected.
There are however several ways that this could be both good and bad, true and false. The original Zen timetable has a Q1 2016 schedule that later got pushed. So the company could have been getting the information from old documents. Another possibility could be that we see Opteron CPUs on the new Zen architecture before we get the consumer FX versions around the holiday sales 2016.
Either way, I am one of them that is excited to see what AMD has cooked up in their labs and how well it will perform compared to the competition.
AMD’s next-gen x86 CPU core architecture, codenamed Zen, has, according to rumour, been fully tested and has “met all expectation [sic]” with no “significant bottlenecks”. A source told WCCFTech that AMD and its partners are “very excited” over Zen, and that the chip could help the company be more “competitive against Intel” following the poor reception to its Fury GPU series this year.
The full text, shared by a commenter calling themselves ‘Lurker’, reads:
“Regarding Zen performance, a guy who worked for AMD (at least his linkedin profile says that) and who, as he claims, worked on designing L2 cache for Zen and K12 said that their focus was to be competitive against Intel. He no longer works there but apparently his old colleague who still works there said Zen chips have already been tested and so far “it has met all expectation” [sic] and they “haven’t found any significant bottlenecks”. Apparently they haven’t finalized the specifications for the clocks and TDP, but their partners in server market are “very excited”.
It’s not much detail, but I think if there was a problem from having only 2 AGUs, it would count as a significant bottleneck. Also this is my first post ever, I just usually lurk here and this is the first time I have something useful to add to the discussion. Please no bully.”
AMD’s 14nm Zen processors will support DDR4 memory and Simultaneous Multi-Threading Support Technology (SMT), allowing a performance increase of up to 40% Instruction Per Clock (IPC). The K12 will be Zen’s ARM equivalent.
The report from WCCFTech’s source regarding the AMD Zen and K12 CPUs has not yet been verified, so should not be taken as fact.
Having moved from PowerPC to Intel’s x86 a little under a decade ago, it looks like Apple may be considering a CPU switch again. This time around, the rumours point o a much less dramatic change, with Intel being dropped in favour of AMD. This is due to AMD’s new Zen architecture which is expected to launch next year, with the new chips expected to improve their IPC by 40% and bring them much closer to Intel’s offerings. A move by Apple to use Zen would serve as a high-profile endorsement and an indicator of Zen’s performance.
One of the biggest reasons Apple may consider a change to AMD is due to the latter’s semi-custom designs. AMD allows their customers to make and choose their own style of APU SoC, allowing creative blends like the chips found in the PS4 and XBox One. While Intel’s move towards better iGPUs can be traced partially to pressure from Apple, AMD would allow more flexibility than Intel currently allows. It would allow Apple, for instance, to order a custom APU paired with HBM in a unique configuration available only to Apple.
Another reason is that Apple stands to save a good chunk of cash as AMD chips ar generally cheaper and the 2nd place chip manufacturer is in a worse bargaining position. While Apple could design their own chips, the need for x86 support still requires either Intel or AMD. Supporting AMD also allows Apple to gain a better bargain from Intel if Apple continues to source chips from the latter.
Apple has started a shift to include more AMD chips recently. Despite using Nvidia for a long period, Apple has moved to include AMD’s Radeon 7970, R9 285 and M370X in their systems. A move to using AMD CPUs and APUs makes sense as long as the performance and efficiency are there. As long as Apple can maintain their computing experience for users with ADM chips, it is unlikely many of Apple’s customers will care.
Thank you WCCFTech for providing us with this information
Despite pretty dismal Q3 2015 results, AMD looks set for a good late 2016 and 2017. AMD has reportedly completed their next generation Zen and K12 designs and have taped them out. This comes after AMD reported that they have taped out a number of FinFET designs during Q3 2015. With both TSMC’s 16nm FF+ and Global Foundries 14nm LPP both using FinFETs, AMD is set to gain good power savings for their next generation.
When Jim Keller left earlier in the year, there were those that had thought that meant Zen was a failure and unfinished due to Keller’s departure. With this information, we can be pretty confident that Zen is largely Keller’s work and was pretty much done by the time he left. K12 on the other hand is based off of the ARMv8 instruction set and will probably do battle against ARM’s own architectures and the likes of Qualcomm’s Kyro.
With Zen already taped out, a Q4 2016 launch for Zen will probably make it in time. This means in about a year from now, we will finally get AMD’s new architecture in our hands. Zen is expected to bring a 40% IPC improvement over AMD’s latest Excavator design, bringing performance competitive to Intel’s designs.
All hopes at AMD are pinned on their new Zen CPU architecture performing well against Intel’s currently superior lineup. While we have seen some leaked block diagrams and claims of 40% IPC improvement come out, other details have been scant. We now have a patch that details pretty much what Zen will look like, at least at the block level. Zen is also expected to bring instruction set parity between the two x86 CPU players.
Overall, Zen may have a total of 10 executions ports. These are meant to feed the integer side 4 ALUs and 2 AGUs, and the FPU consisting of 2 128it FMAC, each with 2 128BIT add and 2 128bit mul. There is a chance that AMD will have fewer execution ports as the ALUs may share ports with the FPU. In order to get the most out of the architecture though, having more execution ports is crucial. With such a wide core, AMD will really let their SMT (Hyper-Threading) stretch its legs with mixed workloads.
Compared to Bulldozer, the ALU count remains the same but being in the same core should give a massive boost to single-threaded performance. Zen does lose out 2 AGUs in the process, but that shouldn’t hurt it too much given our experience with K10. The fact they should be AVX2 compatible should also mean improvements on throughput, at least if the software uses the latest extensions. For the FPU, Zen basically doubles the throughput, which goes a long way towards boosting IPC in floating point heavy workloads. It’s interesting that Zen goes with 128bit units but we’re expecting them to combine to do AVX instructions which should provide parity with Haswell/Skylake. Furthermore, by allowing the FPU to split into 2 128bit units, older instructions may actually run better than on Intel which still can only process 1 128bit instruction despite the execution units 256bit width.
On the instruction decode side of things, Zen cuts things down from Steamroller/Excavator, with only 4 instructions per clock compared to 8. Zen ‘s decoders won’t need to feed 2 cores however as in the Bulldozer design, meaning the real decode rate is the same provided you are running more than 1 core at a time. 4 instructions per clock is also where Intel is currently sitting. For the cache, it looks like we will be seeing a return to the Cat (Jaguar) and K10 design, with 512Kb of L2 per core and 32KB of L1 data, with 32kb of L1 instruction likely as well. While it is a drop, Zen won’t have to feed as many cores and with less cache thrashing it should actually perform better.
With Zen, it really looks like AMD has taken a lot of lessons from K10, Jaguar, Bulldozer and even Intel to create what appears to be a really strong CPU architecture on paper. By combining all of the strong traits from previous and current CPUs, AMD may finally give Intel a run for their money. It’s just too bad we’ll have a year to wait before Zen will arrive. Given Intel’s pace though, Zen should still be plenty competitive in a years time.
Thank youdresdenboy for providing us with this information
Intel has long limited the mainstream platform to 4 physical cores, with 8 threads due only to Hyper-Threading. This has held true for the longest time, with power users who wanted more cores making the jump to LGA 1366 and 2011. According to an Intel CPU engineer profile, Intel will start offering more cores starting with Cannonlake in 2017. Instead of a jump to 6 cores, the alleged Cannonlake SoC will feature up to 8 cores.
While some may say that this is a server processor, that is unlikely due to the SoC designation. That usually points to a mobile environment, with an attached GPU and other dedicated hardware on the same chip. The biggest question is whether or not this core-count increase is going towards mobile devices or the higher performance desktop platform. As we all know, the mobile world ahs become obsessed with core counts so a low power extreme multi-core CPU may be there for marketing. That segment would be better served though by a 10nm shrink of Goldmont from the Atom lineup which should still exist in 2017.
One of the reasons Intel may be moving to more cores on the mainstream platform in 2017, is that software is slowly starting to feed 4 cores quite well. Even with Hyper-Threading, 4 cores may simply not cut it. By pushing more cores to mainstream users, Intel is also pushing software developers who have generally been loath to make their software more-multithreaded. Another reason is that by 10nm, Intel will have enough space and cost savings to allow for 6 and even 8 cores to be cheap enough to produce. 10nm will be 4 times smaller than the 32nm process, allowing for many more chips per wafer.
Finally, 2017 is also the year that AMD will launch their Zen architecture. Set to provide a 40% IPC over current Excavator chips, the new lineup is set to arrive on a 1x nm process and feature up to 8/16 cores + SMT (AMD’s Hyper-Threading) on the mainstream platform.It may be that Intel perceives Zen may be a threat to their mainstream platform and is taking proper precautions. Whatever the reason, 2017 looks to be a really interesting year for CPUs, with major product introductions from both camps.
AMD took a blow this past week after another major departure. While some have taken Jim Keller’s departure to be negative news, we aren’t so sure. Internally though, AMD still seems to be in high spirits and hopes are high for Zen.
Speaking for the Zen design team, engineer Suzanne Plummer is optimistic:
“It is the first time in a very long time that we engineers have been given the total freedom to build a processor from scratch and do the best we can do,” Plummer said. “It is a multi-year project with a really large team. It’s like a marathon effort with some sprints in the middle. The team is working very hard, but they can see the finish line. I guarantee that it will deliver a huge improvement in performance and (low) power consumption over the previous generation.”
Set on a 14nm FinFET process, Zen is expected to boost IPC by 40% over current Excavator designs. While that probably won’t be enough to catch up to Intel’s Skylake, it should place Zen withing striking distance of Ivy Bridge or Haswell. This would be good enough performance to claw back sales provided pricing is competitive. Zen is set to launch in Q4 2016.
Thank you myStatesman for providing us with this information
One of the most renowned members of AMD, Jim Keller is leaving the firm once again. The former Chief Architect of Microprocessor Cores is leaving today to pursue other opportunities. Jim Keller was instrumental in designing the K7 and later K8 CPUs that really put Intel in a tough corner. His work was instrumental in the creation of the x86-64 instruction set that forms the basis of x86 computing today. After leaving 3 years before K8 launched, Keller went on to Apple where he worked on the A4 and A5 SoCs.
Returning to AMD in 2012, Keller is noted as being one of the drivers behind the upcoming Zen microarchitecture. Zen is reportedly set to diverge significantly from the current Bulldozer family and return to a style reminiscent of the K7/K8 cores that Keller designed. Keller was also working on the K12, AMD’s ARMv8 based CPU.
The big question, of course, is how this will impact Zen, which is set to launch in late 2016. Given the long time frames involved in CPU design, most of the work should already be completed, and it would not be surprising if Zen+, the 15% IPC boost over Zen, is also largely finalized. After all, Keller spent a total of 3 years at AMD this time around which is enough time for him to work his magic. Because of this, Keller’s departure likely won’t impact on at the next 2 CPU architectures from AMD.
While some might view Keller due to Zen being a flop, that is unlikely to be the case. Keller did leave 3 years before K8 and the same year K7 launched. With Zen well on its way, Keller is likely moving onto his next project or retirement while AMD continues on iterate based off Zen. Another possibility is that with Zen completed, AMD can no longer afford to keep Keller, though that is a bit pessimistic.
Moving onwards, it means that Zen+’s successor likely won’t see much input from Keller. Those CPUs are still very far away and likely won’t arrive till 2018/2019 so in the near term AMD should be fine. It’s also important to note that AMD does have many more engineers as well as Keller’s team that can also do good work as well. Mark Papermaster, AMD’s Chief Technology Officer, will take over Keller’s duties for now.
Thank you Bloomberg for providing us with this information
Ever since AMD launched the much maligned Bulldozer CPUs, fans have been waiting for chips that would be competitive performance wise with Intel. Next year, however, that is set to change with the launch of the new Zen architecture. Set to provide up to 40% increase in IPC and performance, many are hoping the new chip will provide what it takes to revitalize the CPU market.
While many had hoped that Zen would arrive quickly in 2016, the latest report appears to show that those hopes will be dashed. According to those in the motherboard industry, Zen won’t be launching till Q4 2016, the last 3 months of the year. This will also likely be behind Intel’s Skylake refresh, Kaby Lake and around the time that the enthusiast Skylake-E series launch. With a late 2016 launch, AMD risks introducing a product in between an upgrade cycle. A Q4 launch would also place it behind the back to school season but might make it in time for the 2016 holiday sales.
Sources are speculating that the delayed launch is due to issues with GlobalFoundries’ 14nm process. This process was created in partnership with Samsung so it’s likely Samsung also won;t be able to supply any chips to AMD earlier. GlobalFoundries has been a source of problems for AMD with delays to the 28nm and later 20nm nodes causing either poorer than expected performance or some say even product cancelations.
Thank you DigiTimes for providing us with this information
While AMD has released some details about Zen at their Financial Analyst Day earlier this year, details have still been a bit scant. What we already know is that Zen will have a 40% IPC increase compared to Excavator, bringing AMD’s IPC much closer to Intel’s in one jump. Zen will also support a version of Simultaneous Multithreading (SMT) to support 2 logical processors per core. This will all be bundled on the AM4 platform with DDR4 support and use a FinFET process. Most critically, the CMT or cluster-based threading will be gone and each core will have 2 256bit FPUs and a good number of Integer ALUs.
Today though, we have a rumour that suggests that Zen will bring AMD to instruction set parity with Intel’s Haswell/Broadwell CPUs. With Excavator that launched earlier this year, AMD already caught up partially with AVX 2 which brings 256bit support to integer work, BMI2 and RDRAND for pseudo-random number generation. If Zen is to catch up to Haswell, it will probably add hardware acceleration support to CRC, SHA-256 and RSA algorithms and RDSEED for more pseudo-random number generation. interestingly, there is also suggestions that AMD’s SMT implementation will be compatible with the Intel’s meaning OS’s may not need to be patched, like they did with Bulldozer, to fully support the extra logical processor.
AMD may also support some of the new Skylake instructions like AVX 512 though we will have to wait and see. Part of this is due to the fact that Intel is yet to fully reveal what Skylake supports till IDF later this month. With Intel slipping in a refresh with Kaby Lake in 2016, AMD really has a good chance at a comeback if Zen performs well.
Thank you Fudzilla for providing us with the information