Intel Core i7-12700 + Intel B660 Evaluate

Today we have a combo review of the brand new Intel hardware, starting with our first look at some affordable LGA 1700 motherboards designed for Intel's latest generation Alder Lake processors. These go well with locked-down non-K CPU models like the Core i7-12700 that we have on hand for testing today. Finally, we also have one of the new Intel box coolers, the creatively named RM1, which should be interesting.

Starting with the motherboards, as usual, MSI played their game by quickly sending the MAG B660M Mortar WiFi DDR4 and it looks like this is the only B660 board we have for testing so we're happy to have it. The rest of Intel's partners seem to have gotten into the Christmas spirit, forgetting this launch is happening. Anyway, the B660M Mortar WiFi DDR4 is a great place to start and ideal for showcasing the new Core i7-12700.

Let’s now quickly move to the nine new 65-watt CPUs, minus the F variants, of which there are four. The only Core i9 model is the 12900, and compared to the 12900K it is clocked only 100 MHz lower, which reduces the maximum turbo output from 241 watts to 202 watts, a significant reduction in consumption of 16% for just 2 – 2.5% frequency drop.

There's also a single Core i7 model, the 12700 we're testing today. It is clocked 100 MHz lower than the K-SKU for the P-cores and 200 MHz lower for the E-cores. This reduces the maximum output from 190 watts to 180 watts.

There are three Core i5 models: the 12600, 12500, and 12400 and apparently the only F-SKU will be the 12400F. The 12400 clocks up to 4.4 GHz, the 12500 4.6 GHz and the 12600 4.8 GHz, which is only 100 MHz below the 12600K.

The parts that we can call really new are the Core i3, Pentium and Celeron models. There are two Core i3s, the 12300 and the 12100, both pack 4 P cores plus SMT without E cores with a 12 MB L3 cache. The 12300 clocks at up to 4.4 GHz, the 12100 at 4.3 GHz and this model is available as an F-SKU.

The Pentium Gold G7400 is a dual core with 2 P cores for 4 threads, no E cores, it has a 6 MB L3 cache and runs at 3.7 GHz. After all, the Celeron G6900 is a lower dual-core with 2 P-cores without SMT, i.e. a 2-core / 2-thread processor with a 4 MB L3 cache at 3.4 GHz. This is the Alder Lake processor of the web browser and email editions.

Intel B660 chipset

Intel is launching new affordable chipsets, we're getting the B660, which we're looking at today along with the H670 and H610. In addition to the Z690, there are now four 600 series chipsets, all of which support integrated Intel WiFi 6E, Intel Rapid Storage Technology v19 and PCI Express 4.0.

Focused on the B660, it is very slimmed down compared to the Z690. Of course, CPU overclocking in true Intel form is off the table, but you can at least overclock memory, which is more useful these days anyway. Only x4 DMI Gen 4 lanes are supported by the CPU, which cuts PCIe 4.0 connectivity in half compared to the Z690, but that won't be a problem for most users.

This means that the chipset's PCIe 4.0 lanes have been shortened from 12 to 6, while the PCIe 3.0 lanes have been reduced from 16 to 8 from 10 to 4 and the 5 Gbps ports from 10 to 6. The only one Another feature worth mentioning is PCIe RAID, which is not supported on B660 or B610 boards.

While K-SKU CPUs in standard energy mode maintain maximum turbo performance or the PL2 state indefinitely (PL1 = PL2), the locked processors like the Core i7-12700 work like previous generations, i.e. H. PL2 is a temporary boost state and after a certain time the CPU drops to PL1, which is 65 watts for the 12700.

It's a bit complex and chaotic, it's anything but consumer-friendly. Intel fixed this for the K SKUs, but the locked parts are everywhere. For example, if you install the 12700 on any Z690 motherboard except Asrock entry-level models, it will run in the PL2 state indefinitely, even though it is a locked part. This can also happen with some B660, H670, and H610 boards. For example, the MSI B660M Mortar WiFi DDR4 runs without any power limitation as standard.

This means that the 12700 can use as much power as it wants on the B660M mortar, and normally this should reach a package power of around 180 watts. But if you were to install it on some of the most popular B660 boards, PL2 would likely hit 180 watts for a limited period of time before dropping to PL1 or 65 watts. How this affects the clock rates depends on the workload and for today's test we are testing the 12700 in maximum performance and in the 65 watt specification.

In a nutshell, however, this means that the 12700 could be over 30% faster on some B660 boards depending on the power state. And this gets worse if you're using a K-SKU with a 125W base spec. Anyway, we'll likely get into more on this in the future.

First, let's go through the test system. We passed up testing DDR5, which seems like a waste of time for now, and we know everything we need to know regarding DDR4 vs. DDR5 performance. To test the Core i7-12700, we use the MSI B660M Mortar WiFi DDR4 with 32GB of dual-rank dual-channel DDR4-3200 CL14 memory, the same hardware we use for all of our DDR4 tests and which is usually faster as single-rank DDR4-3800 CL18 memory in terms of performance.

The K-SKU Alder Lake CPUs were tested on the MSI Z690 Tomahawk WiFi DDR4 with the same memory and all boards were updated to the latest BIOS revision. We also updated our Ryzen data using the MSI X570S Tomahawk WiFi.

All game data has been updated for the AM4 and LGA 1700 CPUs with the Resizable BAR enabled. The plan was to do the same with the 10th and 11th generation Intel Core processors, but performance dropped in all cases with ReBAR enabled, so I've disabled this PCI Express feature on these platforms for the time being.

Finally, the final test system notes worth mentioning are the fact that all of the application and gaming data was collected using the AMD Radeon RX 6900 XT graphics card and the operating system of choice was Windows 11. Okay, I think that covers it, let's get into the results …

Benchmarks

Based on the multi-core results from Cinebench R23, we find that the 12700 when combined with a quality cooler on a B660 motherboard can get a score of 21407, which means that it was only 6% slower than the 12700K. Additionally, it could keep up with the Ryzen 9 5900X while crushing the older Core i9-10900K.

When using the RM1 box cooler without power limitation, the 12700 was thermally limited to 19714 points. That is 8% less than the Corsair H170i.

With the 65 W specification enforced, the score dropped to 16017 points, which corresponds to a performance level similar to that of the Ryzen 7 5800X and Core i9-10900K. A very respectable result, considering how little electricity the 12700 uses here.

When it comes to single-core performance, performance and temperature limits are not a problem, as the 12700 with the RM1 box cooler can clock at 4.9 GHz within the 65 W power budget. As a result, all three configurations delivered similar performance and that meant the 12700 was only 3% slower than the 12700K, making it a whopping 16% faster than the Ryzen 9 5900X.

As we move on, we find that the compression performance of the 7-Zip File Manager is pretty much the same regardless of the cooling and power configuration, which suggests that the utilization is not that high.

The same applies to the decompression results, regardless of the configuration used, the 12700 is comparable to the Core i9-10900K and only 4% slower than the K variant.

Next up we have Corona, and we find that with an upgraded cooler, the 12700 can essentially compete with the 12700K and only take 3% longer to handle the workload. With the standard RM1 cooler, but without power limitation, the power is thermally limited, which increases the rendering time to 77 seconds, which means an increase in the completion time by 13%.

With the 65 W specification enforced, the render time is reduced to 88 seconds, which means an increase in render time of almost 30% compared to the H170i configuration. However, we still have a performance comparable to that of the Ryzen 7 5800X and Core i7-11700K.

Adobe Premiere Pro's workload is a mix of single and multi-core tasks. With the H170i liquid cooler, the 12700 was only 2.5% slower than the 12700K, which is a great result. We find that the performance of the box cooler has only been reduced by 2%, a respectable result.

Limiting the 12700's power to the 65W spec resulted in a drop in score to 723 points, a 14% reduction over the H170i configuration, and the 12700 was back on par with the 5800X.

Adobe Photoshop is an application that is largely based on the performance of a single core. Therefore, thermal and power headroom is not an issue as you can maximize a single core within the 65W power budget and the RM1 can easily handle this thermal load. It also means that the 12700 was only 2% slower than the 12700K in this test.

Like Photoshop, Adobe After Effects is largely based on single-core performance. With the H170i, the 12700 was the same as the 12700K, while the limited-power 65-watt specification with the RM1 reduced the score by only 4%, a negligible difference, and so the 12700 did even better than the Ryzen 9 5900X, which is very impressive .

We include the game Factorio in the application benchmarks because we don't measure fps, but updates per second. This automated benchmark calculates the time it takes to run 1000 updates. This is a single-threaded test and it appears to be heavily dependent on cache performance.

Because of the single-core nature of the test, the 12700 performed pretty much the same regardless of the cooler or power level. It was 7% slower than the 12700K but still managed to outperform the Ryzen 9 5900X, which is a great result for Intel's locked down Core i7 part.

When it comes to code compilation performance, the Core i7-12700 is a beast, at least when combined with an upgraded cooler. At 4058 seconds, it was essentially the same as the 12-core Ryzen 9 5900X and lasted only 4% longer than the 12700K.

When thermally limited with the RM1, it took 5% longer to handle the workload, and then 28% longer with the 65W spec. But even here it was able to keep up with the Core i9-11900K in terms of efficiency, an impressive result.

The final application benchmark we'll look at is Blender, and these margins are similar to what we've seen in a number of core-heavy benchmarks. Without any power or temperature limitation, the 12700 corresponds to the 12700K, and if it is thermally limited with the RM1 box cooler, the rendering time increases by 12%.

This time, the 65-watt spec greatly reduced performance and increased render time by a massive 46%, making the 12700 one of the slowest CPUs tested. This is due to the use of AVX commands which blow the power budget and therefore greatly reduce the frequency at which the cores can operate when limited to a packet power of 65 watts.

power consumption

When we measure the total power consumption of the system, we see that the Core i7-12700 with the H170i consumes slightly less power compared to the RM1, even though the liquid cooler allows for higher performance.

We believe this is explained by the fact that the CPU with the box cooler reaches the thermal threshold of just over 100 ° C faster and, since the cooler is unable to efficiently dissipate that much heat, much of it ends up on the board, and this increases VRM temperatures, decreases VRM efficiency, and thus increases overall system utilization.

Gaming benchmarks

Time to check out the gaming performance and we'll start with F1 2021. The 12700 was within a percent of the 12700K at the average frame rate, and averaging almost 400 fps, the performance is more than adequate here.

Even with the 65 W specification enforced, the 12700 averaged 364 fps, which is only a 7% decrease from the H170i configuration. It also meant the 12700 could keep up with the 10900K and 11900K with limited performance.

The Core i7-12700 did well in Rainbow Six Siege with an average of 566 fps, the same level of performance as the 12700K. In 65W mode, the 12700 was 7% slower and while that made the Ryzen 7 5800X faster, we're still talking about over 500 fps on average.

Interestingly, the margins in Watch Dogs: Legion are much smaller and here we only drop 5% with the 65W mode, which limits the output from 12700 to 5900X and 5800X. But with the performance limits removed and the H170i installed, the 12700 was only 2 fps slower than the 12700K, so they're basically the same.

Next we have Shadow of the Tomb Raider and here the 12700 managed to get close to the 12700K with an average of 186 fps and with the installed RM1 box cooler the performance only decreased by 2%, although you could hear the CPU cooler.

Then the performance dropped 7% to 173 fps with the 65 W specification and here the 12700 was between the 5800X and 11900K. Not a bad result given the performance limit.

Riftbreaker is a super CPU intensive game and a new addition to our benchmark battery. This one favors Alder Lake over Zen 3 and even Intel's 11th generation Rocket Lake parts work really well. Again, the 12700 was equivalent to the 12700K on both the H170i and RM1, while the limited-power configuration was only 4% slower, which was equivalent to the 11900K.

Compared to Ryzen, the 12700 was 23% faster than the 5900X, so this isn't a great title for AMD, although overall performance is acceptable.

The 12700 hit the 12700K again when cooling with the H170i, although the RM1 box cooler only lost a few frames, which was basically the same level of performance. Enabling the 65W specification reduced frames by 10% to 189 fps, again the same level of performance as the 11900K.

Next we have Age of Empires IV and this data cannot be compared to the previous results from this game as we are using a completely different test and the resizable BAR is now enabled for Alder Lake and Zen 3 in terms of performance, Alder are Similar to Lake and Zen 3, and since it is primarily a single-core game, the 12700 offers the same performance for all three configurations.

Far Cry 6 is another single core game and this one really does favor Alder Lake's strong single core performance. Even at the 65W power limit, the 12700 was 12% faster than the 5900X. Since the game relies heavily on a single core, the 12700 can still clock very aggressively within this 65W power budget.

Horizon Zero Dawn doesn't make too much use of the CPU, so the 65W spec still delivers great results and allowed the 12700 to displace the 11900K. Although Ryzen takes the cake on this one, the 12700 was only 6% slower, which is the same as the 12700K.

Last we have Cyberpunk 2077, which despite the scaled down quality settings with these high-end CPUs is still largely GPU-limited, with the exception of the Ryzen 7 3700X. The 12700's power limitation reduced performance by 8%, making it even slower than the Core i7-10700K. But without limitation, it was the same as the 5900X and was only a few frames slower than the 12700K.

Before moving on to the average data for 10 games, here is the total system power consumption while running our Cyberpunk 2077 benchmark. Here, too, the RM1 increases the overall system utilization and this is due to the fact that the CPU heats up the VRM, which makes it run less efficiently.

But with a good cooler, the Core i7-12700 matched the 5800X and 10700K in terms of power consumption and then when performance was limited it matched the 3700X, solid results given the performance.

10 games average

Here's a look at the 10-game averages and as expected, the no-capped Core i7-12700 is right there with the K version unlocked, which makes it a bit faster than the Ryzen 9 5900X.

With the 65 W specification enforced, the 12700 achieved an average of 200 fps, which is on par with the 5800X and only 2.5% slower than the 5900X. It was also impressive to see that the 12700 was about 5% faster than the 11900K with the 65W limit.

Intel box cooler: Laminar RM1

Another new component that we are testing in this review is the Intel Laminar RM1 box cooler. There's also a larger RH1 version that comes with the Core i9 parts locked and then the simpler RS1 version that comes with Pentium and Celeron parts. They're all 65w coolers technically, so the RM1 is designed to only work with the 12700 if the 65W spec is enforced.

In terms of heat sink design, the RM1 doesn't differ that much from previous box coolers that were equipped with the Kupferslug, that is, the Kupferslug is much larger and this helps to reduce the overall weight from 226 grams to 346 grams, a massive difference of 53%. The cooler also looks better and has a more robust mounting system. Overall, this is a much better product, but it's only rated for 65 watts, so don't expect anything amazing.

Take a quick look at the cooling performance with the RM1 box cooler at the 65W specification in a room at 21 ° C with the components inside the Corsair 500D installed with the doors closed. We're looking at an operating temperature of 76 ° C under an all-core workload in Cinebench R23, looping for an hour, and that's a respectable result.

When the performance limits are lifted, the CPU core temperature quickly reaches 100 ° C and this leads to throttling. The CPU frequency depends on the workload and of course we already looked at the application performance under thermal limitation with the RM1 cooler.

To avoid throttling, a simple tower-style air cooler is sufficient, we only see slightly more power consumption than the Ryzen 9 5900X and significantly less than the Core i7-11700K.

MSI B660 VRM

At this point we've seen how the Core i7-12700 performs on the MSI B660M Mortar WiFi DDR4, but how does the board perform? We don't mean storage, USB or audio power, but the all important VRM. Was the board about to cook itself or are you good at running anything up to the 12900K on it?

To find out, I tossed it into our Corsair Corsair iCUE 7000X VRM test system and let Cinebench R23 loop for an hour. To record the temperatures we use a digital thermometer with type K thermocouples and report the peak temperature to the back board. After all, I don't report a Delta T via Ambient, instead I keep a room temperature of 21 degrees and ensure a constant ambient temperature by positioning a thermocouple next to the test system.

At the 12900K, the B660M mortar peaked at just 79 ° C, which is a great result and means that even in a 30 ° C room, the peak PCB temperature is well below 100 ° C and you usually don't see any VRM throttling unless the temperature rises above 100 ° C. with 125C is dangerous.

This result means the mortar isn't that much hotter than entry-level Z690 boards and is worlds better than the Asrock Z690 Phantom Gaming 4, a board I bought to test out but haven't covered in detail yet, so think me, spoiler warning. In terms of performance, the Mortar got the most out of the 12900K after the hour-long stress test and was 7% faster than the Asrock Z690 Phantom Gaming 4. So a great result for a mid-range B660 that I'm expecting Panel.

What we learned

That was our look at the Intel Core i7-12700, MSI's B660M Mortar and the new RM1 cooler from Intel. The box cooler is certainly an upgrade, it's not a game changer, and we expect the vast majority of you to ditch it in favor of a more capable cooler. Even a $ 20 tower-style cooler will be an upgrade.

For those working on the 65 watt spec, it works and doesn't look terrible. It's actually a nice solution given the size constraints, but won't be particularly appealing or useful to most enthusiasts.

The Core i7-12700, on the other hand, is extremely responsive and will likely be a first choice for many PC manufacturers. When fully unleashed with a good cooler, the 12700 was only ~ 7% slower than the 12700K in applications. So for those who aren't into overclocking, this should make the 12700 a much better deal.

When it comes to gaming, the 12700 and 12700K are largely indistinguishable. The locked portion was at most 7% slower, but typically the margin was less than a few percent and less than a percent overall separated the two in the 10 games tested. So, in terms of value, the 12700 will make a lot more sense for those who aren't into CPU overclocking.

Speaking of prices, at the time of this writing we don't have official pricing information from Intel, but if we look at the 11th generation 11700 series we can make some estimates. The MSRP is $ 400 for the 11700K, $ 375 for the 11700KF, $ 320 for the $ 11700, and $ 300 for the 11700F. The 11700 is 20% cheaper than the 11700K. Assuming we see a similar margin on the 12700K and 12700, the 12700 would retail at around $ 330, but Intel could be less aggressive on pricing for this generation given its competitiveness.

Say Intel hits $ 330 for the 12700 or 12700F, and a board like the B660M Mortar goes for about $ 170, which is an extraordinary $ 500 package and less than what you get for the Ryzen Pay 9 5900X, which currently costs $ 540.

Meanwhile, the 5800X costs $ 370, plus $ 150 for a decent, comparable-quality B550 board, and that combo is a bit more expensive for an inferior package. So will Intel be so aggressive about pricing? Cant tell at the moment, but we'll have an idea shortly after this review goes live, and a really good idea in the next few weeks.

If one thing is certain, AMD will be wiped out in the value segment with these 13 new locked CPUs from Intel. Assuming the Blue Team can meet the demand, and assuming B660 boards are competitive, parts like the 12400F are going to be particularly deadly and we want to see that shortly.

The bottom line is that the Core i7-12700, and especially the F variant, seem like a great option for high-end gaming.

Purchase abbreviation:
  • Intel Core i7-12700 on Amazon
  • Intel Core i7-12700K at Amazon
  • Intel Core i5-12600K at Amazon
  • MSI Z690 Tomahawk WiFi DDR4 on Amazon
  • MSI Z690-A Pro WiFi DDR4 at Amazon
  • Asus Prime Z690-P D4 on Amazon
  • AMD Ryzen 9 5900X on Amazon
  • AMD Ryzen 7 5800X on Amazon

Leave a Reply

Your email address will not be published. Required fields are marked *