Intel is introducing new hybrid architecture to its CPU with Alder Lake. However, the hybrid CPU idea is not new. After the Apple M1 chip, Intel is the second one to use a hybrid design in computing chips.
The announcement date of 12th gen Alder Lake is October 27, and the launch date, including reviews and sales, is November 4.
High-end SKUs will launch in Core i9, Core i7, and Core i5.
Core i9-12900K and 12900KF
Core i7-12700K and 12700KF
Core i5-12600K and 12600KF
Intel Z690 chipset
Pat Gelsinger is very confident about the performance of the new Alder Lake. Moreover, rumors and leaks also suggested that Alder Lake is far ahead of the Ryzen 5000 CPU.
Intel Alder Lake will End AMD’s Superiority
Gelsinger said in an exclusive interview with CRN:- “AMD has done a solid job over the last couple of years. We won’t dismiss them of the good work that they’ve done, but that’s over with Alder Lake and Sapphire Rapids,”
After, Pat Gelsinger was appointed as the CEO of Intel Company. The company has made many changes for the benefit. Intel still holds up to 80% of the market share, including consumers, servers, etc.
Moreover, the company decided to rename its process node. Previously called 10nm Enhanced SuperFin is now Intel 7, and 7nm is now Intel 5, etc. We all know that Intel was struggling with the 10nm process yields, but the 10nm process is more advanced and equal to TSMC 7nm.
Intel and Microsoft collaborated to optimize the new Windows 11 operating system according to the hybrid design. The OS optimization is necessary so that the application or any task uses the right cores every time.
You wouldn’t want your games accidentally run on efficient cores (E-Cores) rather than performance cores (P-Cores). And the low power background task started utilizing a performance core. To avoid that OS scheduler needs to run properly, and Intel also added a Thread Director mechanism to smooth up things.
We have recently seen some side effects of Windows 11 OS on AMD Ryzen CPUs. Windows 11 reduced up to 20% performance of Ryzen CPU. However, AMD is working with Microsoft to resolve this issue, and sooner or later, it will be resolved.
The impact on Ryzen CPU shows that some major changes have been made to the Windows OS that would benefit Alder lake hybrid CPU.
On the other hand, AMD is not sitting ideal. They are working on 3D V-cache that they showcased in Computex 2021 that helps to increase up to 15% performance. Will probably be released in Q1,2022.
Many factors will affect the market share of both companies. Intel’s aggressive price and performance will be a game-changer. AMD, since the Ryzen 5000 launch, started charging a higher premium than Intel.
Undoubtedly, Intel has made a substantial amount of changes, both inside and outside the company. But defeating AMD would be not easy.
Intel CEO Hopes To Get Back Apple Business
In an interview on “Axios on HBO,” Gelsinger said – he doesn’t blame Apple for dropping Intel and deciding to make its own chips.
“Apple decided they could do a better chip themselves than we could,” he said. “And, you know, they did a pretty good job. So what I have to do is create a better chip than they can do themselves. I would hope to win back this piece of their business, as well as many other pieces of business, over time.”
Apple M1 chip uses ARM architecture, whereas Intel uses the x86 architecture. With the M1 chip, Apple has achieved more than desired performance for the MAC platform, at least more than Intel used to give.
Also, inducting the M1 chip has helped to reduce the device sizes by making it more compact. On the software side, Apple has built several things to enable a seamless transition from Intel x86 to the M1 ARM chip.
Once a company switches to its in-house chip, it’s nearly impossible to go back. Beyond that, there’s no reason for Apple to switch back. With the recent M1 Pro and M1 Max, the performance has gone up even more.
Today, a variety of PC cases are available that fit in almost every scenario. Generally, we denote the case size with Tower like Full-Tower, Mid-Tower, and small form factor cases like Micro-ATX, Mini-ITX. Also, all cases have different expansion features for HDD, SSD, PCIe expansion slots.
However, not every PC case is suitable for everyone. Different users have different component requirements, and it is necessary to select PC cases accordingly.
Here, we have discussed some significant points that will help in getting the proper case for your PC, like airflow, future component expansion, looks, design, price, etc.
As we said above, a variety of sizes are available in cases. And yes! When it comes to features, size matters. Whether you want lots of fans and radiator space, drive expansion, custom liquid cooling, bigger GPU, etc.
Full-Tower PC Case
Full-Tower cases are huge thus take more space on the table. It supports E-ATX, ATX, Micro ATX, and Mini ITX motherboards. Also, it provides plenty of more room for custom water cooling, radiators, and fans.
Drive expansion is simple here as it has lots of expansion options. And no restriction for graphics card sizes and numbers.
Consider a full-tower PC case only if you plan to use an e-ATX motherboard, heavy custom water cooling, multiple GPU setups (like three or four).
Mid-Tower PC Case
Mid-Tower cases are evergreen for almost any type of PC building. These cases are compact than Full-Tower and most usual. It supports regular motherboard sizes like ATX, Micro ATX, and Mini ITX.
It provides ample space for fans, radiators, drive expansion (including SSD and HDD), cable management, and more. For AIO water cooling, Mid-size cases are more than enough.
Most of the Mid-size cases supports high profile graphics card too. But it’s recommended checking the size for compatibility.
Micro ATX & Mini ITX PC case
The small form factor cases are decent for users who have a space crunch, especially. However, anyone can use it according to their needs. Generally, cases like this provide less flexibility with components.
Fewer options for fans and limited drive expansion. AIO liquid cooler can be used but depend upon the feature availability. In addition, low-profile and limited-sized GPUs fit. Check the GPU size compatibility with the case.
Mini-ITX case airflow isn’t as good as Mid-Tower and Full-Tower cases. Micro-ATX PC case fits Micro-ATX and Mini-ITX motherboard, whereas Mini-ITX case only supports Mini-ITX motherboard. The big CPU cooler won’t fit in either.
Airflow matters most
You don’t want your components to suffocate due to low airflow. There are cases in the market which has poor airflow and ventilation. Don’t go for looks and design only. I know it matters but not at the cost of compromising airflow.
Search PC case according to what size CPU cooler, Air/Liquid cooler, you want to use. Custom liquid cooler generally required Full or Mid-Tower case.
Adequate ventilation is required to pull cool air in and push warm out of the case. Cases that offer tight airflow clearance, especially at the front, should be avoided. Some cases even draw air from the bottom.
Despite good ventilation, poor fan orientation could ruin the airflow. It’s necessary to configure the fans in the correct order to get optimal thermal performance. Few fan orientation examples are below.
Look for a dust filter as it reduces the accumulation of dust inside the case. Dust on components can cause multiple issues like heating, reduces airflow, etc. However, don’t forget to clean your case and dust filter regularly, even if you use a dust filter.
Price:- Make a Decent Budget.
It is important to choose a decent budget for your PC case. Most cheap cases don’t offer a better build quality, adequate expansions, ventilated chassis, more pre-installed fans, etc.
At least select a case with two or more fans at the front and one exhaust fan at the back.
Up to ₹5000 ($50):- Many cases fit under this price point that offers overall good value. For cheap builds, it’s OK to spend less on a PC case. A decent number of drives and fans can be installed.
Between ₹5001 – ₹15000:- Expensive build with high-end CPUs need great cooling solutions. Cases in this price bracket won’t disappoint in features like liquid cooling, fans, expansion, large CPU cooler, big size GPU support, etc.
The cases with this price are more compatible with liquid cooling, offer better cable management, provides a better look with performance, more pre-installed fans, etc.
Decide the features you are looking to have in your case and then search. More expensive cases, above ₹15000, are also available in the market. A good Full-Tower case is generally pricey, and the most affordable is the Mid-Towers.
However, cases type, such as Full or Mid-Towers, don’t decide the prices. Many feature-rich Mid-Towers are expensive than Full-Tower and vice versa.
Features to look
Sufficient SSD/HDD Mounting
Sure, we get some SSD/HDD mounts in every case. The problem appears when you have several of them, and no slots or mounts remain.
Good airflow and dust filter
Proper airflow is very crucial for any PC you ever build. The heat-generating components are everywhere inside the case, like motherboard, GPU, CPU, etc. Good airflow prevents components from reaching high temperatures. Lower temperatures help to output more performance.
Dust easily gets accumulate inside in high airflow cases. To avoid this, try to get a PC case with a dust filter. Although a dust filter does reduce airflow by some amount, it’s worth it if you don’t clean the PC very often.
Of course, you don’t want your PC to look ugly due to poor cable management, that too if your case has a tempered glass side panel. Although, most of the case nowadays supports good cable hiding and routing place.
Just find out which case is providing it better while meeting the other requirements.
Support big air cooler
Keeping low temperatures is difficult on powerful multi-core CPUs (Core i9, Core i7, Ryzen 9), especially with small-size cooler. And for any reason, you want to avoid liquid coolers. If you are planning to use one of these CPUs, then look for a case that supports the huge size air cooler.
Supports liquid cooling
Other than air coolers, liquid coolers are the far best option to keep the CPU cool efficiently. Different sized radiators are available in an AIO liquid cooler like 120MM, 140MM, 240MM, 280MM, 360MM, etc. Also, custom liquid cooling needs extra space for a reservoir, tubes, etc.
Look for the radiator sizes are supported by the case or not, which you are planning. And not all AIO liquid cooler supported case supports custom loop liquid cooler fittings.
Support bigger GPUs
Modern graphics cards are getting bigger in length and width. Many of us are familiar with three slot cards, although two slot cards are common. Similarly, three fan graphics not only gives good thermals but also are bigger.
Keep in mind that not every case supports the larger size graphics card. Check out the case specification to check the maximum GPU length it supports.
Modern front-panel connectivity
Front panel connectivity ports provide the facility to connect USB, audio jack, etc. Some cases also offer RGB control buttons.
More number of advanced ports is always better. Check for USB port numbers and versions like USB 2.0, USB 3.0, USB 3.2, etc. It is a big pain if you have to go to the back of the case to connect high-speed devices.
Lighting in a PC case makes it look amazing and the center of attraction. The use of RGB lighting significantly increased in all the PC components.
If you like RGB lights and want to build a full RGB PC, choose a case accordingly.
Vertical GPU mount
A regular GPU mounting method almost hides the graphics card look. The graphics card’s fan, RGB lighting, look, and designs are all visible in the vertical GPU mount.
However, it’s better to fit the graphics card like the traditional way (horizontal). Cases do not provide better airflow to the GPU when it is mounted vertically. Graphics card fans are so close to the side panel that interferes with the smooth airflow.
Vertical GPU also blocks access to the bottom of the motherboard.
Cases with tempered glass are becoming common these days. Having a tempered glass PC case is a must for those who plan on building a full RGB PC. Undoubtedly, the RGB illumination and tempered glass look magnificent and add essential characters.
I, personally, like the PSU a lot because it hides most things and gives a clean look. It greatly helps in cable management. If you bought a non-modular PSU, you could easily hide unused cable under the PSU shroud.
A PSU shroud also helps to enhance the overall look of the PC case.
Overall demand for the graphics card is insane, and prices are skyrocketing. And During this GPU or silicon shortage, new Nvidia GPU SKUs would make the condition worst.
The new desktop RTX Super will use the same silicon die used in the current RTX 3000 like GA102, GA104, GA106. However, the GA103 die is in rumor.
Instead of fulfilling the current stocks, it’s not a good idea to launch any new Nvidia GPU. You, as user and buyer, can understand and relate it very well.
GA102 – RTX 3080, RTX 3080 Ti, RTX 3090
GA104 – RTX 3070 Ti, RTX 3070, RTX 3060 Ti.
GA106 – RTX 3060
Imagine you waited long for RTX 3080 prices to come down in your budget. Despite allocating more GA102 GPUs for RTX 3080, Nvidia announced a new RTX 3080 Super, powered by GA102. What would you think? Here Nvidia would impact not only the stocks of both graphics cards but also the prices.
The prices have already gone up, much above the MSRP. Adding a new “Super Variant” at a higher MSRP will increase the price significantly. I know performance would increase with “Super” cards, but it’s not as important as price and availability.
Moreover, Nvidia already launched its “Ti variant” of the RTX 3000 series at Computex 2021 (except RTX 3090 Ti). Despite having the “Ti” and low availability of GPUs, it makes no sense of a new “Super” variant, at least for now.
Furthermore, the GPU die of RTX 3000 Super can be used to make more current RTX 3000. Unless you find your desire card at a reasonable price, “RTX Super” would be an utterly stupid move by Nvidia.
Nvidia RTX 2060 Relaunching
More surprisingly, Nvidia is planning to relaunch RTX 2060 with 12 GB of VRAM in early 2022 as a stop-gap solution. That indicates that the Nvidia GPU shortage will persist in the year 2022 also.
Nvidia itself, delaying the low-budget RTX 3050 series card. That shows the silicon supply and manufacturing issues. So, how could the company launch the new Super variant?
Furthermore, the global silicon shortage won’t go away anytime soon, and demand for Ampere cards is already high. In addition, crypto miners are absorbing graphics cards, which has also led to stock shortages.
HP leaked RTX 3080 Super Nvidia GPU
Recently, a leak surfaced from hp that shows the RTX 3080 Super card in the all-in-one PC. Technically, All-in-one PCs use Mobile/Laptop GPU, not desktop. There is the possibility that the new Super variant may come in Mobile GPU. But it is less likely to happen in the desktop card segment.
The laptop GPU die’s size is smaller than the desktop. Hence the chip yield per wafer is high. In addition, the clock speed is much lower, so the performance is less than that of the desktop.
Many leaks are present, showing details of RTX 3060 Super, RTX 3070 Super, RTX 3080 Super, and RTX 3090 Super. Nvidia could have launched it if the world wasn’t struggling with the global silicon shortage and GPU shortage.
Every foundry, especially TSMC, in the world working at full capacity to meet the demand. Building new fabrication factories is in progress, but it will take time.
A new mid-range AMD RX 6600 graphics card is arriving very soon. It will use Navi 23 GPU and target the 1080p gaming segment. The performance and price are unknown as if it will align with or go down with the RTX 3060.
VideoCardz revealed that the new card is supposedly launching in mid-October. Leaked embargo information shows that the Press/Reviewers will get the card by October 4, not before that.
After that, on October 13, 2021, at 9 AM EDT, the review will go live across all platforms. However, some information is allowed to share before, like photos, availability dates, AMD approved content.
Radeon RX 6600 XT GPUs are marketed as “for Ultimate 1080p gaming”. Maybe the Non-XT RX 6600 will be for “Better 1080p gaming”.
AMD RX 6600 Specs
Radeon Rx 6600 will feature an RDNA 2 based Navi 23 GPU with 28 computer units (CU) (32 CU in RX 6600 XT). Supports 8 GB GDDR6 memory, PCIe 4.0, and has a 32 MB infinity cache.
AMD skipped the reference design card for RX 6600.
Things are skeptical here, whether AMD will keep its price closer to the RTX 3060’s $329 MSRP or under $300.
At present, no AMD and Nvidia cards are available at or near MSRP. As was the case with the RX 6600 XT, the first shipment had decent prices, but the second sale went crazy with the prices.
Everyone is wondering about Intel’s ARC GPU lineup and performance output. Intel is planning to target the low end and mid-range segment graphics card market.
On Architectural day, 2021, Intel released information on the Alchemist (DG2) ARC gaming graphics card. The new ARC GPUs will launch in Q1, 2022.
As shown in the leaked slide, the proposed DG2 lineup would have two SOCs. The SOC1 with TDP of 175W- 225W, and SOC 2 with a TDP of 75W.
The mid-range SOC1 would go against RTX 3070, RTX 3060 Ti, RTX 3060, RX 6700 XT, RX 6600 XT. And SOC2 would target low-end $100 – $200 GPUs like GTX 1650 Super
The low-end segment has been completely ignored as there are no new generation cards available by both Nvidia and AMD. Moreover, the majority of gamers buy low-end and mid-range cards ($200 – $500, approx up to ₹45,000). Like GTX 1650 Super, GTX 1660, RTX 3060, RTX 3060 Ti, RX 6600 XT, RTX 3070.
We all know that the GTX 1060 is so popular due to its price and performance. In Nvidia, the XX50, XX60 series graphics cards are considered to be the best selling.
The current graphics card market is messed up by mining, silicon shortage, etc. Even a low-end card is selling at a mid-range price.
If Intel targets the most popular segment of GPU, that too with aggressive prices, then no one can stop Intel ARC GPU to become a gamer’s choice. Intel needs to do something new, and for now, the new thing is the price, nothing else.
Nvidia and AMD are already offering advanced features like Ray tracing, DLSS, FSR, etc. Nvidia’s offering is top-notch whether we talk about ray tracing or DLSS. In that situation, Intel can do a price battle to gain a mass-market share.
It’s better to build a product for a large number of gamers rather than high-end for a few. In 2022, we may get more budget segment graphics cards.
For over a decade, we’ve seen the same big aluminium block stock coolers from Intel. From Core i9 to Pentium, every Non-K 65W CPU contains the same aluminium stock cooler.
With the upcoming 12th gen Alder Lake CPU the stock cooler will probably change for the first time. In recent few generations, Intel has only done a few cosmetic changes to its stock cooler, like a black colour heatsink.
As per the leak, new stock coolers are in a plan with the 12th gen Alder Lake CPU. Alder Lake will support LGA 1700 socket that needs a new cooler anyway.
Three thermal solutions leaks are surfaced for 65W Alder lake-S CPU.
Intel Laminar RH1 – for Core i9
Intel Laminar RM1 – for Core i7 to i3
Intel Laminar RS1 – for Pentium and Celeron
Intel Laminar RH1 – for Alder Lake Core i9
The high-end RH1 cooler has a screw retention mechanism, as shown, that surely needs a back plat. Also, the fan is illuminated blue, which means it supports an LED fan, but no confirmation of RGB lights. Specially designed for Core i9.
No copper heat pipes are visible like AMD used to offer with RGB Wraith Prism and MAX Cooler.
Laminar RM1, RS1 – for Core i7 to Celeron
The medium RM1 cooler is designed to work with Core i7 to i3. This cooler has a clip retention mechanism and no led or RGB fan supported. But a Blue illumination ring around the fan is shown.
A small RS1 cooler is designed to work with Pentium to Celeron CPU. This also has a clip retention mechanism.
New stock coolers are maybe or may not be a part of Alder Lake. We all have to wait until next year when Intel will launch all SKUs of Alder Lake, including the non-K variants. In 2021 we will see only the high-end K and KF variants of CPUs.
All new cooler features the bottom face fan and no major structural and design change. So, we might not see any performance change.
AMD offers the best stock coolers with their processors. And it looks like Intel is rethinking the cooler now.
As the launch of the Intel 12th Gen Alder Lake CPU is near, valuable leaks are surfacing. Probably, the top-tier CPU and Z690 motherboard will launch this year and the remaining SKUs in 2022.
The Intel 12th Gen Alder Lake-S desktop core processors will include “K” and “non-K” variants. Moreover, hybrid Alder Lake architecture processors will be available from desktop to ultra-thin notebook segments.
Intel 12th Gen Alder Lake Release Date
Wccftech said – Our sources have confirmed that the Intel 12th Gen Core processor lineup along with the Z690 motherboards will be readily available on 19th November.
As per leaks, At the Intel Innovation Event on October 27-28, the company would announce the 12th Gen Alder Lake lineup, prices, specification, general benchmark, etc.
After few weeks of the announcement, the launch date will be around mid-November (19th November). Furthermore, Windows 11 is scheduled to launch on 5th October, as the OS is specially optimized to support Alder Lake’s hybrid architecture.
In addition, the top expensive CPU like Core i9-12900K, Core i7-12700K, and Core i5-12600K may launch first with the Z690 chipset motherboard. And the other H-series and B-series chipset motherboards are rumoured to be available in early 2022. A New 1700 socket motherboard is required to run the new CPU.
Alder Lake supports both DDR4 and DDR5 RAM. So, the 600 series chipset motherboard will support DDR5, but DDR4 support is also planned. For example, an expensive motherboard may get DDR5 support, while a cheap one may get DDR4.
Also, momomo_us tweeted the full SKU list from Core i9 to Core i3 last week.
Alder Lake-S Desktop CPU Price
No doubt, Alder Lake is Intel’s next big architectural change. A huge performance leap is a must for Intel to counter AMD Ryzen 5000 and 6000 CPUs. Other than performance, the prices will also need to be very competitive and aggressive.
Leaks suggest that the upcoming Alder Lake-S desktop CPUs will be expensive than 11th gen Rocket lake.
Core i9-12900K – $705 – ₹62,300 (Inc. 18% GST)
Core i9-12900KF – $674 – ₹59,600 (Inc. 18% GST)
Core i7-12700K – $495 – ₹43,800 (Inc. 18% GST)
Core i7-12700KF – $464 – ₹41,000 (Inc. 18% GST)
Core i5-12600K – $343 – ₹30,000 (Inc. 18% GST)
Core i5-12600KF – $312 – ₹27,600 (Inc. 18% GST)
India prices are including 18% GST.
These prices may surprise you a bit. Prices seem more expensive if we compare Intel to Intel. However, there is no point in comparing the previous 11th generation Rocket Lake with the price as the 12th generation has a higher core count, new hybrid architecture etc.
Similarly, Ryzen 9 5950X, a 16 core 32 thread CPU, retailing at $749 (₹76,000) and Ryzen 9 5900X, a 12 core 24 thread CPU, retailing at $609 (₹48,700).
The overall performance of the CPU will also determine the price. It remains to be seen how the combination of Alder lake’s P-core and E-core will work to beat AMD.
It looks like Nvidia and AMD are planning to introduce new GPUs. Probably the new GPU will be a refresh or advance version of the current generation. As rumour suggests that the new card could be RTX 3090 Super and RX 6900 XTX.
Many gamers are either waiting for the GPU price decrease or a lower budget new generation card launch. Both the cases are looking impossible as the cryptos again rising. In this puzzling situation, these companies are thinking to launch new high-end GPUs.
Nvidia RTX 3090 Super Rumour
On the other hand, new high-end RTX 3090 Super leaks are roaring. It has 10752 Cuda cores, 21 Gbps memory speed and a massive 400+W TGP(Total Graphics Power). The leak says the TGP is greater than equal to 450W, but I think it’s a mistake, he means less than equal to 450W. The upcoming GPU is power-hungry, and it is not surprising as this will be a top-tier card in the RTX family. Maybe the overall TGP will be around 400W.
Furthermore, Nvidia may offer 1 TB/s memory bandwidth for the first time on the RTX 30 series card using 21Gbps memory and 384 Bit interface. And the die will possibly be the same GA 102. However, we may not see NV-Link in 3090 Super.
Before the launch of the RTX 30 series, it was a rumour that the RTX 3090 would feature 21 Gbps memory, but it launched with 19.5 Gbps.
Nvidia had saved some memory space for future launches according to market and competition.
RTX 3080 Super is also in rumour, and it is possible as the specification gap is huge between RTX 3080 and RTX 3080 Ti.
RTX 3080 Super is also in rumour, and it is possible as there is a massive specification gap between RTX 3080 and RTX 3080 Ti. Nvidia can easily slot one more card between them as RTX 3080 Super.
AMD RX 6900 XTX Rumour
AMD is also gearing up to counter the Nvidia launch. The Radeon RX 6900 XTX will be based on Navi 21 with 18 Gbps memory and 24.93 TFlOPS.
The RX 6900 XT currently has two variations, a standard variant that features the Navi 21 XT GPU and a binned variant featuring the Navi 21 XTXH GPU. Both of these variants are known as Radeon RX 6900 XT.
AMD may be planning to use binned Navi 21 XTXH GPU die with faster memory. This step will help to counter the current Nvidia’s RTX 3090.
Leaks suggest the RX 6900 XTX will offer 24.93 TFLOPS Compute Performance, which is roughly 8% more than the 23.04 TFLOPS of RX 6900 XT. Moreover, the bost frequency of RX 6900 XTX will be between 2300 – 2400 MHz, which is again higher than 6900 XT 2250 MHz.
Radeon RX 6900 XTX can surpass the exiting RTX 3090 in raster performance if all rumours are true.
As rumours, the launch date of RTX 30 Series Super and RX 6900 XTX will be in 2021. No exact month is published anywhere. But AMD launch is more sceptical than Nvidia. No concrete launch information about AMD card.
Maybe both companies have more capable performance chips. So instead of wasting them on exiting cards, they thought of building a new SKU and generate more revenue.
Global chip manufacturer, TSMC, has notified its customers, like AMD, Nvidia, Qualcomm, Apple, about the price hike. Taiwan semiconductor manufacturing company will increase the production cost of most of the process technologies. Undoubtedly, the companies will transfer the increased cost to the customer in their products like CPU, GPU, Smartphones.
TSMC plans to increase prices by about 10% for its 7nm (N7) or less advanced process nodes, whereas a 20% price increase for 16nm or thicker process nodes. Most of the modern silicon chips use TSMC 7nm nodes and later will use TSMC 5nm (N5). And 10% production cost increase will significantly increase the overall customer value.
As reported by TSMC, its 49% revenue is coming from 7nm (31%) and 5nm (18%), and 25% revenue is from 16nm (14%) and 28nm (11%).
Most of the silicon is for smartphones and computer parts. That means cost increase would impact these two platforms heavily.
Prices are already not normal due to silicon shortages and other factors. In addition to this, a hike in the prices of TSMC, the world’s largest foundry, may make the situation worse.
Currently, silicon shortage is every sector that uses silicon chips. From computers to automobiles, it’s a global chip shortage. Gradually, things are going back on track, and stocks are normalizing. TSMC and other silicon manufacturers have started building new fabs to fulfil the upcoming demands.
Manufacturing new silicon fabrication facilities required billions of dollars. There may be many reasons for the price hike, but more revenue, new fabs, chip shortage are part of it.
Impact on CPU, GPU prices
In computing, many companies use TSMC foundry for chips like AMD, Nvidia, etc. Intel planned to use TSMC N6 process node for its upcoming Intel ARC GPU. However, Intel uses its own foundries for CPUs. Nvidia also uses Samsung foundry, but TSMC is in priority.
AMD majorly uses a TSMC 7nm process node for the CPUs and GPU and later will use 5nm. Ryzen 5000 CPU and the upcoming 6000 CPU may cost more than we expected. Also, AMD has planned to use 3D stacking to give extra cache in the CPU.
The mobile SoC market for TSMC is huge. Smartphone consumption is increasing rapidly globally, so is the number of SoCs. Major chip designers like Qualcomm, MediaTek, Apple rely on TSMC. And most of them are using advance 7nm or lower process nodes.
It remains to be seen how much the increased production cost affects the price of CPU, GPU, mobile SoC, etc. Either the companies will absorb the price hike in them or pass it on to the customer. No company would want to hurt its profit margins.
In Architecture day 2021, Intel has released much information on Alder Lake hybrid architecture. Intel Alder Lake CPU will consist of new Performance (P-core) and Efficient (E-core) x86 core microarchitecture.
Surprisingly, the Efficient core of Alder lake gives more performance than the Skylake core. The Skylake architecture is not too old as Intel used it up to 10th gen Comet lake (optimized Skylake).
Once was the powerful cores are now becoming the Efficient cores of modern Alder Lake chip.
Compared with the Skylake single-thread performance, the Efficient core achieves 40% more performance while consuming 40% less power. However, most probably, Intel has been used the 6th gen Skylake cores number for the comparison.
Whereas, in multi-core performance, four Efficient cores achieve 80% more performance while consuming 80% less power than two core four thread Skylake. The hyperthreading is disabled in all Efficient-core.
Unfortunately, no detailed benchmark reports have got from the company. And all these numbers are claimed by the company after using SPECrate 2017 benchmark tool. Independent testing will show the real capabilities of the CPU.
Efficient cores may play an important role in the mobile segment. Intel Alder Lake will be introduced across desktop, mobile and ultra-mobile platforms. In the mobile segment, the E-cores are more than the P-cores.
Better utilization of Efficient cores will probably give an advantage to Intel-powered laptops. The E-cores not only will help to gain performance but also in reduced power consumption. And the CPU overall power consumption is very important for the laptop’s battery life.
The Performance core (P-core) will deliver the important CPU performance. It consists of a new Matrix Engine to accelerate AI workload. Also, it has a new PM controller for power budget management.
Intel claimed – the Alder lake P-core gives an average of 19% more performance than 11th gen Rocket Lake. Not only that, Intel targeted to reduce all types of latency. The Performance-core is the highest performing CPU core Intel has ever built and pushes the limits of low latency and single-threaded application performance.
Some rumours suggest that the Alder Lake Core i9-12900K has surpassed the Ryzen 9 5950X in Cinebench R20. Here, Ryzen 9 5950X is a 16 core 32 thread CPU whereas, i9-12900K will have 8 Performance cores and 8 Efficient cores. This means the hybrid architecture impacted the performance, and Efficient cores are not weak.
All the leaks surfaced is of an engineering sample of the CPU, not a retail unit. And Windows 11 is still developing under insiders preview. Later, as the scheduler becomes more mature, we may notice some significant performance differences. So retail units of CPU and matured Windows 11 can give more output than we expected.
Intel Thread Director
In order for Performance-cores and Efficient-cores to work seamlessly with the operating system, Intel has developed an improved scheduling technology called Intel Thread Director. And it is built directly into the hardware.
The Thread Director is a must-have in the chip. Without it, the OS scheduling would be a nightmare. Because each application or program will have to land on the correct core when launched. Moreover, any problem while allocating cores to the application may ruin the user experience.
For instance, if games and rendering apps accidentally lands to E-Core, it will degrade performance significantly.
It is also important to switch the program from p-core to e-core and vice versa, depending on the performance required.
Intel Alder Lake Spec
Undoubtedly, the Intel Alder Lake CPU is a huge leap forward. It is the first performance hybrid architecture with P-core and E-core, built on Intel 7 process node (Previously, Intel 10nm Enhanced SuperFin). The P-cores are hyperthreaded, whereas the E-cores lacks hyperthreading.
Alder Lake will feature the new Xe graphics and modern connectivity like Thunderbolt 4, Wi-Fi 6E. Support for both DDR4-3200 and DDR5-4800 gives a wider choice of RAM usage. Still, the cache is lower than what AMD used to offer. Supports up to 30 MB LL cache.
This will be the first platform with PCI-e Gen 5 support. The PCI-e 5 has double the bandwidth of PCIe 4. Hence using more PCIe devices won’t bottleneck the lanes. New 1700 socket motherboard required to run Alder lake desktop CPU.
The power consumption will be higher than expected as the rated TDP is up to 125W. Maybe the power utilization is less with laptops and no restrictions on desktops.
Indeed the hybrid architectures are outperforming. Previously, the Apple M1 chip showed us some surprising numbers and is now used in Macs. Likewise, we are expecting the same result with Intel Alder Lake.
Furthermore, Intel Alder Lake is the last resort to counter AMD current and upcoming CPUs. AMD planned a 3D stack to increase cache in the future CPU that has already been demonstrated in Computex 2021.
On paper, Intel Alder lake looks promising but more details are yet to be released.