Some Thoughts About the Intel Core i9-9900KS Processor

Introduction

Intel has released the Intel Core i9-9900KS Special Edition mainstream desktop processor that was originally announced back on May 26, 2019 at the Computex trade show. This was the same day that AMD revealed the details about the 7nm AMD Ryzen 3000 series mainstream desktop processors. The preview announcement of the Core i9-9900KS was seen at the time as an attempt to upstage AMD’s announcement of Ryzen 3000 series.

Now, roughly five months later, Intel has actually released the Core i9-9900KS, which they are touting as “the world’s best gaming processor”. Is this true, and should you want one of these processors for your next gaming rig?

See the source image

Figure 1: Intel Core i9-9900KS Packaging (from PCMag.com)

Specifications

Let’s start with the main specifications for this new processor. It is a 14nm, 8C/16T “Coffee Lake” processor with a base clock speed of 4.00 GHz and a Max Turbo speed of 5.00 GHz. It has a 16 MB L3 cache, 16 PCIe 3.0 lanes, and a 127W TDP rating. It uses an LGA 1151 socket and will work in existing 300 series motherboards with a BIOS update. It includes integrated Intel UHD Graphics 630, but does not have a stock CPU cooler in the box. The recommended customer price is $513.00. The warranty is only one year instead of the normal three years that Intel usually offers on most of its other processors.

This new processor is essentially a specially binned version of the existing Intel Core i9-9900K processor (which was released in Q4 2018) that will let you run all of the eight cores at 5.00 GHz without any manual overclocking. One challenge for this new processor is that the existing Core i9-9900K can usually be overclocked to 5.00 GHz on all cores, depending on whether you were lucky in the silicon lottery (meaning you got a “good” sample), your CPU cooler, and your motherboard and BIOS settings. Individual motherboard vendors can choose to use features like Multi-Core Enhancement (MCE), and they can also decide how they want to regulate power usage and Turbo duration. These factors can have a huge effect on benchmark results and real-world performance. Keep in mind that you will want/need a high quality, fairly expensive CPU cooler for an Intel Core i9-9900KS. This might cost anywhere from $75-$200.

Gaming Performance

Intel claims this is the absolute best processor for gaming, but is this actually true? This depends on what type of gaming you plan to do, especially the screen resolution you will be playing at. It makes a big difference whether you game at 1080P (1920 x 1080) or lower; or whether you game at 2K (2560 x 1440) or 4K (3840 x 2160). If you game at 1080P or lower, then your single-threaded CPU performance is usually your bottleneck for getting high frames per second (FPS), assuming you have a good enough video card such that the video card is not the bottleneck. You can use a relatively low performance video card and still get high FPS performance when you are running at 1080P or lower resolution.

Once you go above a certain level of FPS (depending on whether you have a high refresh rate, low latency monitor and whether your monitor has G-Sync or FreeSync), getting even more FPS is not going to make any noticeable difference in your gaming experience. Once you go above 1080P gaming, the video card becomes the bottleneck, unless you have an extremely slow processor. As long as the processor meets a certain relatively low level of performance, your FPS performance in 2K and higher gaming is gated by your video card performance.

What this means is that if your main priority is the absolute highest FPS performance at 1080P or lower, then yes, the Intel Core i9-9900KS is the world’s best gaming processor. If that use case is not your main priority, there are other less expensive choices, and also other similar cost choices that will work much better for many other scenarios.

Other Processor Choices

If you are an Intel fan, you could choose an Intel Core i9-9900K (with the right motherboard and BIOS settings), and very likely get the same level of 1080P gaming performance as a 9900KS. You would probably save $50-$100 by doing this, depending on the actual street prices of both processors. You could also choose a less expensive 8C/8T Intel Core i7-9700K processor and probably save $200-$250 and still get nearly the same level of 1080P gaming performance as a 9900KS. In either case, you could spend those savings on a better video card, or keep the savings yourself. If you are thinking about building a new Intel desktop system, I would urge you to wait until November/December, since there are pretty strong rumors that Intel is going to reduce the prices of their older existing CPUs by perhaps $50 or more.

What about using an AMD processor? One excellent choice for gaming and general purpose usage is the 7nm 8C/16T AMD Ryzen 7 3700X, which is currently selling for $319 (including a pretty good stock Wraith Prism CPU cooler in the box) at Micro Center. The Ryzen 7 3700X is pretty competitive with these Intel processors for gaming, depending on the game and the resolution. It is going to be much better than the Intel Core i7-9700K for general purpose computing since it has 16 threads instead of only 8. You could use the money savings (vs. a Core i9-9900KS) for a better video card, more RAM, or better storage.

You could also step up to a 12C/24T AMD Ryzen 9 3900X (if you can find one) or wait a few weeks and get a 16C/32T AMD Ryzen 9 3950X (which may also be in short supply when it is released). The 3900X is $499 and the 3950X will be $749. These processors will also be pretty competitive with the Intel processors for gaming, depending on the game and the resolution. They will be much better than any mainstream Intel processor for workstation performance because of their higher core and thread counts and PCIe 4.0 support. Speaking of that, the upcoming Ryzen 9 3950X is very likely to be better than the new $979 18C/36T Intel Core i9-10980XE Cascade Lake-X HEDT processor for many workstation workloads.

Frankly, the initial written reviews for the Intel Core i9-9900KS have been pretty brutal:

The Intel Core i9-9900KS Review: The 5 GHz Consumer Special

The Intel Core i9-9900KS Special Edition Review: 5.0 GHz on All the Cores, All the Time

Intel Core i9-9900KS Special Edition Review: More power, less point

There are also many YouTube reviews that have been even more negative about the Intel Core i9-9900KS. Here are a few:

Intel Core i9-9900KS Review, Winner of 2019’s Most Boring CPU Award
Intel is selling BINNED 9900Ks! Core i9-9900KS Unboxing
$513 5GHz Special Edition CPU – Intel 9900KS Review
Intel i9-9900KS Review: Overclocking, Power, & Gaming CPU Benchmarks
INTEL i9 9900KS Release! REVIEW & OVERCLOCK to 5.4 GHz!
Intel i9-9900KS Marketing: Rushed and Hilarious!
Well this is awkward…My golden sample 9900K BEAT the 9900KS

Conclusion

Does all of this mean that the Intel Core i9-9900KS is a “bad” processor? No, absolutely not. If 1080P gaming with the highest FPS performance is your main concern, then this is a great processor. As long as you have a high quality Z390 motherboard with good VRMs, a high quality CPU cooler, and a good enough graphics card, you will be very happy.

If you game at higher resolutions, or do other things besides gaming, you have other choices that are much more affordable, such as an AMD Ryzen 7 3700X processor. For the same amount or slightly more money, you can get much better general purpose and workstation performance with a higher core count AMD Ryzen 9 3900X or 3950X processor.

What are the implications for the overall mainstream desktop CPU market? Well currently, Intel is losing a lot of both market-share and mind-share with their current offerings. The Core i9-9900KS is a pretty weak response to the AMD Ryzen 3000 series outside of one particular narrow use case. Intel’s next short-term move is likely to be price reductions, which they have never really had to do in the past. Intel has milked the 2015-vintage 14nm Skylake architecture (and its Kaby Lake, Cannon Lake, and Coffee Lake derivatives) about as far as they can.

Intel will eventually have a better response, probably in the late 2020/2021 time frame. By then, AMD should have the next generation Zen 3 processors. If you enjoy computer hardware, this is a great time to watch what is happening in the industry! 




Glenn’s Tech Insights For October 11, 2019

(Glenn’s Tech Insights… used to be part of our bi-weekly newsletter but we decided to make it a regular blog post instead so it can get more visibility. It covers interesting new hardware and software developments that are generally relevant for SQL Server). It also can just be technically related items that I find interesting.

Microsoft SQL Server Cumulative Updates

Microsoft has recently released (and then removed) SQL Server 2016 SP2 CU9. There was nothing wrong with the actual payload of SQL Server 2016 SP2 CU9, but there was a problem if you tried to uninstall it. To be fair, uninstalling a CU doesn’t happen that often, but if you want/need to do it, it should work properly. It was quickly replaced with SQL Server 2016 SP2 CU10, which is identical to CU9, except that the uninstall issue has been fixed.  Since SQL Server 2016 SP1 is no longer supported, there was no CU for that branch.

Microsoft also released SQL Server 2017 CU17 on October 8, 2019. This is Build 14.0.3238.1, with 34 public hotfixes. So far, there are no reported issues with this CU. I wrote more about this CU here. Despite some recent stumbles by Microsoft, I am still a big proponent of trying to keep your SQL Server instances as up to date as possible. That does not mean throwing a new CU into Production the day it is released, but it also does not mean avoiding patching SQL Server indefinitely either.

Intel Cascade Lake-X HEDT Processors

On October 7, 2019, Intel released a new line of high end desktop (HEDT) processors for the existing X299 chipset. This is the 14nm Core i9-10900 Series that is meant for HEDT and workstation usage on the LGA2066 platform on existing X299 and new X299X motherboards. There are four SKUs in this release, ranging from 10C/20T up to 18C/36T.

These processors have slightly higher base and Turbo Boost clock speeds than the previous Skylake-X HEDT processors, with support for up to 256GB of DDR4-2933 memory. They also have several other minor improvements compared to Skylake-X. Overall, the majority of the enthusiast tech community seems to be pretty underwhelmed by this product family.

The big news with this release is a rather dramatic price reduction compared to Skylake-X, especially on the higher core count SKUs. For example, the new 18C/36T Core i9-10980XE has a launch price of $979.00 compared the previous 18C/36T Core i9-9980XE that went for about $2000.00.

New Intel Core I9 X Series Refresh SKUs LGA 2066

Figure 1: Intel Core i9-10900 Series SKUs

These price cuts seem to be a pretty obvious response to what AMD has been doing over the past year (and what they are going to release in November). AMD is due to release the 7nm 16C/32T mainstream desktop Ryzen 9 3950X for $750.00 in November, along with the 7nm 3rd Generation Threadripper processors. The 3rd Generation Threadripper processors are rumored to have between 24C/48T at launch, up to possibly 64C/128T later. There are also strong rumors of new TRX40 and TRX80 chipsets that may have additional memory channels and more PCIe 4.0 lanes compared to 2nd Generation Threadripper. We don’t know clock speeds or pricing yet.

This is another example of why viable competition between Intel and AMD is good for the market and for consumers. AMD seems to be winning a lot of battles in different market segments lately, but Intel has many talented engineers and lots of resources that they can throw into the fight. Intel will eventually have a better response than drastic price cuts, so this will be an interesting fight over the next couple of years.




T-SQL Tuesday #119 Changing Your Mind

This post is a response to this month’s T-SQL Tuesday #119 prompt by Alex Yates.  T-SQL Tuesday is a way for the SQL Server Community to share ideas about different database and professional topics every month. This month’s topic asks us to write about something in your IT career that you have changed your mind about. What was your original opinion? Why did you believe that? What do you believe now? Why did you change your mind?.

MJ-t-sql-Tuesday


Introduction

I’ve been writing and presenting about SQL Server hardware for many years now. I actually wrote a book called SQL Server Hardware back in 2011, which is probably the only book about server hardware from a SQL Server perspective. During most of my coverage of SQL Server hardware related topics, I have been a big proponent of using Intel processors for SQL Server workloads. I even used to make joking comments during some of my presentations about how AMD probably had a contract out on me, because I gave such negative opinions about their processors for SQL Server usage. There were valid reasons why I had a negative opinion of AMD processors though.

Why I Used to Only Recommend Intel Processors

After the introduction of the 45nm Intel Nehalem microarchitecture in 2008, Intel was completely dominant compared to AMD when it came to single-threaded CPU performance for server processors. Intel was also successfully executing their Tick-Tock release cadence, where they released a new CPU microarchitecture roughly every two years (a Tock) and then introduced a manufacturing process shrink on the same basic microarchitecture (with minor improvements) the following year (a Tick).

This Tick-Tock release cycle made it easier to plan when to push for a hardware upgrade or a complete data platform refresh. Figure 1 shows the Intel Server Processor Family Tree from 2008 to the present. The Tick-Tock model was in place from 2007 until early 2016. During this period, Intel captured over 99% of the overall x86 server CPU market, with many server vendors discontinuing their AMD-based server models. The situation was so bad that Microsoft offered a 25% license cost discount if you used qualifying AMD processors with Biztalk Server and SQL Server 2012 or 2014 in a document called the Core Factor Table.

Intel Server Processor Family Tree

Figure 1: Intel Server Processor Family Tree


When SQL Server 2012 was released on April 1, 2012, it brought with it a move to core-based licensing for SQL Server, which replaced the old processor-based licensing that was used in previous versions of SQL Server. This licensing change made it very important to do some careful analysis before selecting the exact processor(s) for a database server or virtualization host that was going to be running SQL Server VMs.

This licensing change by Microsoft happened shortly after the introduction of the ill-fated 32nm AMD Bulldozer microarchitecture, which featured up to 16 physical cores in the Opteron 6200 series. Unfortunately, these AMD processors had very poor single-threaded performance and high power usage to go with their high SQL Server licensing costs. This was not a good combination, hence the Microsoft Core Factor Table.

For licensing purposes, Microsoft didn’t care whether you had a fast processor core or a slow processor core, the license price per core was exactly the same. Knowing this, a smart database professional could purposely select the fastest processor SKU at a given core count, and also try to select a lower core count (but faster) processor SKU in order to get the most performance per core and in order to minimize their SQL Server core license costs.

If you did this properly, it was pretty easy to save many tens of thousands of dollars on SQL Server core license costs on a two-socket server, which would more than offset the hardware cost of of a typical two-socket database server. Using a frequency-optimized processor SKU would also deliver better single-threaded CPU performance and more memory bandwidth than an entry level processor SKU at the same core count.

Why I Now Recommend You Seriously Consider AMD Processors

For Opteron server processors, AMD was stuck with the Bulldozer microarchitecture (and minor improvements with the newer Piledriver and Excavator microarchitectures) until the release of the completely new Zen architecture in 2017. During this period, Intel continued to completely dominate the server CPU market, with no meaningful competition from AMD, except on the low-end. As often happens with a lack of competition, the market leader became complacent over time, to the point where AMD was able to catch up and actually surpass Intel in many areas. As a result, AMD is starting to regain some market share in the server CPU market.

Intel has badly stumbled over the past 4-5 years as they have tried to move from a 14nm manufacturing process to a 10nm manufacturing process. This has forced them to abandon the old Tick-Tock release cycle, and it has also lead to a longer product cycle overall.

This has been accompanied by a very noticeable reduction in generational performance increases since Broadwell-EP, as shown in Figure 2. These numbers are estimated TPC-E scores for a two-socket server with two, eight-core processors, using the fastest eight-core processor from each generation. Based on this, we have seen an extremely small increase in performance over the last four years.

Generational Xeon Performance Increases

Figure 2: Generational Intel Xeon Processor Performance Increases


Intel Security Vulnerabilities

Intel has also had to deal with multiple processor security vulnerabilities. These include Spectre, Meltdown, Foreshadow and their variants, along with newer exploits such as Zombieload. The latest CPU security vulnerability I have heard of is NetCAT, which only affects Intel processors.

Generally speaking, modern Intel processors are more vulnerable to these types of attacks than modern AMD processors are. The required fixes for these vulnerabilities also have a more negative effect on performance for Intel processors than on AMD processors, especially since AMD processors are not affected by many of these exploits. Microsoft’s current SQL Server specific guidance about this subject is here.

On August 7, 2019, AMD finally unveiled their new 7nm EPYC 7002 Series of server processors, formerly code-named “Rome” at the AMD EPYC Horizon Event in San Francisco. This is the second generation EPYC server processor that uses the same Zen 2 architecture as the AMD Ryzen 3000 Series desktop processors. These new processors are socket compatible with the previous generation 14nm AMD EPYC 7001 Series processors, so they will work in existing model servers (with a BIOS update). Despite that, you will need a new model server to be able to use PCIe 4.0 support from the newer processors. The major server vendors like Dell/EMC and HPE have already released new server models that let you fully leverage the new AMD EPYC 7002 series processors.

AMD is not slowing the pace of innovation, since they are on track to release the Zen 3 based “Milan” server processors in Q3/Q4 of 2020. These will have IPC improvements, better L3 cache performance, and probably increased clock speeds.

AMD EPYC 7000 Series Roadmap

Figure 3: AMD EPYC 7000 Series Roadmap


Conclusion

I have previously written about why the AMD EPYC 7002 Series processors are going to be significant for SQL Server. The 7nm EPYC 7002 series has higher memory density and capacity, higher memory speed and bandwidth, more PCIe bandwidth, PCIe 4.0 support, and much lower pricing than 14nm Intel Cascade Lake-SP processors.

A one-socket AMD server will be a superior replacement for many legacy two-socket Intel servers, while a two-socket AMD server will also be a superior replacement for many legacy four-socket Intel servers. I think we will see AMD’s server CPU market share go up to 10-15% over the next 12-18 months. We may also see more aggressive pricing from Intel as a result (which has already happened with the new Intel Cascade Lake-X HEDT and Xeon W workstation processors).

I think an AMD platform is a viable choice for many SQL Server workloads, so I have changed my mind compared to what I thought in the past.