Have you ever noticed how paper and flyer-based computer sales ads focus on processor speed? The truth is, speed is only a single factor in overall CPU performance! Find out how fast your CPU really is.
How fast is your CPU? A note on GHz-based CPU ads
Paper and flyer-based computer ads seem to focus only on processor speed in GHz. You could almost call it sneaky sales oildepending on how much focus there is on the speed of the CPU as a single factor of interest.
The truth is, actual processor speed is only a single part of overall CPU performance. Most important is the actual and overall performance of the CPU. It is true that a given CPU can run at different speeds, and a higher speed will generally result in a higher number of transactions per second (or millisecond).
However, if you compare two completely different processors and say, “The 4 GHz processor must be so much faster than the 1
An important factor in addition to processor speed, for example, is the number of threads. You can imagine a thread-like tube. If you have a single thread processor, it will have a single tube through which the water can flow.
If you have a dual-threaded CPU, it will now have two tubes. It gets a little more complicated when you include Hyper-Threading – which in simple terms would be a “double pipe, shared to some extent” as opposed to a “fully self-contained bong”.
The two main CPU brands are Intel and AMD. For example, modern Intel CPUs often have 16 threads, although these are often 8 real threads and 8 HT (Hyper threads). In other words, 16 pipes, but in reality it is like 8 times two pipes where the two pipes are shared to some extent.
You can see that immediately a fast (in terms of GHz only) CPU, with only a single thread can then be slower a slower (in terms of GHz only) CPU, with many threads. Still, that’s not all there is to it.
Next is caching. A CPU has multiple levels of cache. Often these are called L1, L2, L3 etc. (Level 1, Level 2, Level 3, etc.) Caches. You can think of this as buckets filling and overfilling. The L1 cache is super fast and the closest to the core.
If the required information does not fit in L1, L2 is used (which may be larger but slower). When L2 is full, L3 is used, and (if your CPU has only 3 cache levels) when L3 is full, the data is moved to main memory, which is generally much slower than Lx caches.
You can see that immediately a fast (in terms of GHz only) CPU, with a small L1 cache can then be slower a slower (in terms of GHz only) CPU, with a large L1 cache. Still, that’s not all there is to it.
Bus / channel width is another consideration. The bus is a mechanism by which the CPU communicates with devices and main memory. Think of it like a multi-lane highway: a 64-lane highway will move traffic much faster than a 16-lane highway. While the real world has yet to emerge is a 64-lane highway, but your computer may already have one!
Finally, one can have multiple physical CPUs in one machine, or in other words “multiple actual processors on the motherboard”. Where the term multi-core (multiple cores) can be used in this context, it is somewhat misleading. A CPU can have two cores baked in one silicon chip, using only one CPU socket on the motherboard, or there can be two cores in two physical silicon chips, occupying two CPU slots on a motherboard with two CPUs slots.
Using a dual slot CPU motherboard will tend to the CPU processing power of your workstation. It will almost always never be a true ‘double achievement’ as several other interaction issues come into play. There is potential for separate memory channels, shared memory channels, different ways of connecting to devices and other system components, etc. all of which contribute to the actual performance of a dual (or more) CPU system.
All of this shows us a great need for a better way to evaluate CPU performance, both before clicking the ‘Buy Now’ button and afterwards – to make sure your CPU performance is as you expect. A benchmark if you want, and preferably one that is not influenced too much by environmental and setup differences.
With knowledge comes responsibility; it would be a good idea to avoid returning a CPU to the manufacturer or supplier only because it performs a little less than a published number. Slight differences in manufacturing and raw material can result in a slightly faster or slower version of the same CPU.
Think of it like buying apples at the supermarket; not every apple is the perfect Granny Smith you saw in the TV ad. And no matter how advanced the technology or manufacturing processes are, after all, a CPU is made from the materials of this Earth.
Let’s see how fast our CPU really is.
A reliable measure
What if there was a list where you could quickly look up a particular processor make and model? and what would provide you with a simple numerical benchmark for almost every CPU out there in the world? Such a list exists. Not only that, more than 1 million CPUs were benchmarked, and for a small $ 29 single user license fee, you can buy their performance testing software to test your own PC.
I personally have never found it necessary to buy their performance testing solution: when researching which CPU to use for a build, I can simply look up the performance of each CPU under consideration and compare the ratings (the ‘CPU Mark’ review, more on this later) with each other.
And to evaluate a particular CPU in any workstation or server solution, I can simply search again for that particular make and model number and get the same CPU marking assessment to evaluate against others. But if I ever wanted to actually benchmark a machine, it would be their performance testing solution that I would buy and buy.
After a while you also get a ‘feel’ for what numbers are fast CPUs (generally 7000 and above), and what are slow CPUs (3-4k range and below). Fast is a bit arbitrary here and I should define it better: I consider fast CPUs to be CPUs where one will not notice any slowing down of the machine when performing many tasks at once.
You often see this ‘waiting for the computer’ syndrome especially on slow laptops with slow hard drives (mechanical, rotating hard drives). As an aside, such slow machines often benefit the most from an upgrade from the HDD to SDD (solid state drive).
For a modern, fast workstation, you’ll want to look at at least an 8000 CPU Mark. And higher for servers. Server-based CPUs are also benchmarked and on the same list. Without further ado, here’s the full list on the CPU Benchmark website:
Do you want to find your CPU?
I would be too. If you are using a Linux based workstation, you can easily find out what your CPU make and model is by running the following at your terminal / shell command prompt (a shortcut for this is normally available by ‘shell’ or ‘terminal ‘in your desktop activities or similar bar):
cat /proc/cpuinfo | grep -Ei 'model name' model name : Intel(R) Core(TM) i9-9900K CPU @ 3.60GHz model name : Intel(R) Core(TM) i9-9900K CPU @ 3.60GHz model name : Intel(R) Core(TM) i9-9900K CPU @ 3.60GHz ...
Copy the model number (i9-9900K in this case), and hit CTRL + f on the CPU Benchmark list page referenced above. Then press CTRL + v to insert the same, and your browser will immediately jump to a line with this text.
You will find the exact CPU close to this line, but you will have to match a little more: the search found ‘Intel Core i9-9900KS @ 4.00GHz’ which does not match (note the extra ‘S’ and the different GHz speed) . Press “V” (to next occurrence) or “>” (forward to next time) on your browser search bar to find the next match for ‘i9-9900K’.
Next game shown is ‘Intel Core i9-9900KF @ 3.60GHz’. Almost, but not exactly the same (note the extra ‘F’). Another ‘next’ click to take us to the correct CPU:
Sometimes one has to keep a close eye on the letter specifications in the model name and also make sure that the speed in GHz matches. Note that the output shown in `/ proc / cpuinfo` (from experience it’s a bit difficult to remember when you need it, so take a moment to remember the exact file name and location), shows you the default model speed that the CPU is reporting, not the actual current speed.
To see the actual speed, you can use inxi (
sudo apt install inxi for Ubuntu / Mint or
sudo yum install inxi for RedHat / Fedora) and run
inxi -C (CPU info):
inxi -C CPU: Topology: 8-Core model: Intel Core i9-9900K bits: 64 type: MT MCP L2 cache: 16.0 MiB Speed: 4700 MHz min/max: 800/5000 MHz Core speeds (MHz): 1: 4700 2: 4700 3: 4700 4: 4700 5: 4700 6: 4700 7: 4700 8: 4700 9: 4700 10: 4700 11: 4700 12: 4700 13: 4700 14: 4700 15: 4700 16: 4700
Note how the 16 threads operate at 4.7 GHz (4700). Also note how you can see the minimum and maximum supported speeds (0.8 GHz to 5 GHz). Finally, of note, note the number of cores versus threads, note how there are 8 cores, yet 16 threads.
This is an Intel CPU with HT (HyperThreading) so it has 16 ‘bongs’, which is back to our earlier discussion, but they are shared 2-by-2 and somewhat limited across 8 cores. Another easy way to see how many threads a CPU has can be done by:
cat /proc/cpuinfo | grep -Ei 'model name' | wc -l 16
And we can also check how many cores (all under one physical chip) are on our PC:
cat /proc/cpuinfo | grep 'core id' | sort -u core id : 0 core id : 1 core id : 2 core id : 3 core id : 4 core id : 5 core id : 6 core id : 7
sort -u displays only unique entries. There are (before they are uniquely sorted) a total of 16 entries, but 2-by-2 will show them the same core – 16 threads, 8 cores, HT 2-by-2. And finally, we can even see how many physical CPU chips are installed and running:
$ cat /proc/cpuinfo | grep 'physical id' | sort -u physical id : 0
If this had been a dual CPU system there would have been two lines (second with physical id ‘1’) present!
Other performance factors and alternative measurements
As we have seen, many factors influence the speed of a CPU. If we think a little bigger now, we can also see how many other environmental factors can affect the overall speed of our system. For example, memory speed (DDR3 vs DDR4), virtualization overhead for virtualized remote servers, network bandwidth, and a host of other factors. While they do not directly affect the capacity or ultimate benchmark of a CPU, they all directly affect the overall performance of a CPU.
So keep in mind that when maximizing CPU performance, your CPU may run much faster than your memory, disk, network, etc. Think of it like installing a Ferrari engine in a Jeep car. The car is allowed to drive (if you have a skilled mechanic!), But never use the real performance engine, because the rest of the car is designed for a different purpose.
In addition to https://www.cpubenchmark.net/cpu_list.php, there is also https://cpu.userbenchmark.com/ which is another great website to explore before buying a new CPU. In addition, both websites also offer other benchmarks (GPU, Drive, RAM) that can come in handy in further configuring other performance factors of your new machine!
Check out the performance before buying that next CPU! Before committing to a dedicated server for your business, check out the performance … It’s an easy 2 minute step to take and it often saves you a hefty amount (note the last two columns of the CPU- list, which includes the ‘CPU value’ (higher is better) and ‘price’ per CPU), and this will also regularly optimize the performance of your new workstation or server.
The higher the ‘CPU Mark’ (sometimes referred to as CPU-B), the faster the CPU, all related things (i.e. bus bandwidth, L1-L2-L3-Lx cache, number of threads and cores, etc.) are taken into account .