- Apple has released its new M4 chip with the iPad Pro at the “Let Loose” event.
- In comparison to the Apple M2, the M4 offers 1.5x faster CPU performance and 4x improvements in GPU performance.
- The Neural Engine has also seen a significant improvement in the M4 chip for AI tasks.
Apple has finally unveiled its new family of Apple Silicon chipset at the “Let Loose” event. The Apple M4 has been introduced with the new OLED iPad Pro . The iPad Pro lineup has jumped directly from Apple M2 to Apple M4. So in this article, we compare the Apple M4 and M2 to evaluate their performance difference in CPU, GPU, and Neural Engine. We have also discussed memory bandwidth and other AI-related improvements on the M4. On that note, let’s begin.
We have updated the frequency specification of Apple M4 based on the recently surfaced Geekbench listing. In addition, we have added the Geekbench score of the Apple M4 chipset into our analysis.
Apple M4 vs M2: Specs Comparison
Here is the specs comparison between Apple M4 and M2. Take a glance to understand the difference between the two chipsets.
Apple M4 vs M2: CPU
Starting with the CPU first, the Apple M4 is developed on TSMC’s 2nd-gen 3nm process node which is said to be more efficient. It packs a whopping 28 billion transistors. The Apple M4 CPU for iPad Pro is available in two configurations: 9 CPU cores and 10 CPU cores.
You get 3 performance and 6 efficiency cores in the first variant and 4 performance and 6 efficiency cores in the second variant. And the Apple M4 CPU is clocked at a mighty 4.4GHz, much higher than Apple M2 and M3 CPU clock speeds.
Image Courtesy: Apple
In addition to that, Apple has packed next-gen ML accelerators in the Apple M4 CPU. It can help in AI-related tasks, working in tandem with the Neural Engine.
On the other hand, the Apple M2 CPU is fabricated on TSMC’s 5nm process node and it packs close to 20 billion transistors. It has 8 CPU cores, consisting of 4 performance and 4 efficiency cores, and clocked at 3.49GHz. According to Apple, the M4 CPU is up to 50% faster (1.5x) than the M2 CPU.
If we consider the leaked Geekbench score of Apple M4 , it seems Apple M4 really packs a powerful CPU. It outranks Apple M2 by a huge margin.
Image Courtesy: Geekbench
Apple M4 vs M2: GPU
In the GPU department, Apple has worked on the next-gen GPU architecture which was introduced with the M3 chip. The Apple M4 comes with 10 GPU cores and supports features like dynamic caching for better utilization of the GPU.
Image Courtesy: Apple
In addition to that, for the first time, the iPad lineup gets hardware-accelerated Ray Tracing and HW-accelerated Mesh Shading, thanks to the powerful Apple M4 GPU. Ray Tracing can deliver realistic shadows and reflections in games for an immersive gameplay experience.
The Apple M2 also has 10 GPU cores but it doesn’t feature these capabilities. Apple claims that the Apple M4 GPU delivers up to 4x better performance than the M2 GPU. And in terms of performance-per-watt, the Apple M4 GPU can deliver the peak performance of M2 while consuming half the power.
Apple M4 vs M2: Neural Engine
Coming to the Neural Engine, well, Apple has significantly improved the 16-core Neural Engine on the M4. Despite packing the same number of cores, it’s much more faster and efficient than the M2 Neural Engine. The M4 Neural Engine can deliver up to 38 trillion operations per second (TOPS) whereas the older M2 Neural Engine can only muster up to 15.8 TOPS.
Image Courtesy: Apple
And working with the ML accelerators in the CPU and next-gen GPU, the M4 Neural Engine offers an unbelievable level of performance. Apple says it can run powerful Generative AI and Diffusion models locally on the device.
Apple M4 vs M2: Memory Bandwidth and Others
Finally, in terms of memory bandwidth, Apple has packed a slightly faster memory on the Apple M4. It can deliver speeds up to 120 GBps and the bandwidth is the same across 8GB and 16GB variants. The Apple M2 features LPDDR5 memory that offers speeds of up to 100 GBps on the base variant. The Cupertino giant has not disclosed the memory type for the Apple M4 chip.
Image Courtesy: Apple
Apart from that, the Media Engine on the Apple M4 brings HW-accelerated Av1 decode support for watching high-resolution videos efficiently. It still doesn’t have HW-accelerated AV1 encoding though. And you do get video codec support for ProRes, H.264, and HEVC on both chipsets.
The Verdict
Apple says the M4 chipset is an “ outrageously powerful chip ” and much faster than the M2 chipset while drawing less power. In comparison to the M2 chip, the Apple M4 indeed looks like a big upgrade . You get 1.5x better CPU performance; 4x better GPU performance, 2x improvement in Neural Engine performance, and slightly faster memory speed.
Even in comparison to the Apple M3 , the performance difference is quite substantial. We are interested in power consumption figures of the M4 as it looks like a highly-efficient chipset . Once we get our hands on the new iPad Pro, we will surely test it out and share the benchmark scores from the Apple M4 chip. So stay tuned with us!
Passionate about Windows, ChromeOS, Android, security and privacy issues. Have a penchant to solve everyday computing problems.
Add new comment
Name
Email ID
Δ
- Apple and Google have officially confirmed their team-up for the next-gen Siri AI.
- The Cupertino Giant will be using Google’s Gemini for a more personalized Siri model as well as Apple Intelligence features.
- We can expect the next-gen Siri to come out with iOS 26.4, sometime in March or April.
Apple has officially confirmed joining forces with Google to use its Gemini AI model to power the next-generation Siri. It will offer a more personalized experience and will be coming out with the iOS 26.4 update. Apple also plans to leverage Gemini’s capabilities for other Apple Intelligence features as well, later down the line.
The Next-Gen Siri will be powered by Google’s Gemini AI
Apple officially confirmed that it will be partnering with Google in a statement to CNBC . Here’s what it stated, “After careful evaluation, we determined that Google’s technology provides the most capable foundation for Apple Foundation Models, and we’re excited about the innovative new experiences it will unlock for our users.”
Later, Google also shared a post on X confirming the tie-up, “Apple and Google have entered into a multi-year collaboration under which the next generation of Apple Foundation Models will be based on Google’s Gemini models and cloud technology. These models will help power future Apple Intelligence features, including a more personalized Siri coming this year.”
Image Credit: X/@NewsFromGoogle
Both statements clearly mention that the Cupertino Giant will be using Gemini to power its assistant Siri. This was already rumored, as Apple’s attempts to acquire Perplexity went nowhere. With Gemini, Siri will get a major AI update . It will be able to handle more nuanced conversations and provide better results. Something long-time Apple users have been asking for years.
The next-gen Siri will arrive with the iOS 26.4 update, which will launch sometime in March or April. And it is only going to be available for Apple Intelligence-supported devices .
Something else worth noting is how Google’s statement mentions that Gemini will power Apple Intelligence features. This leads us to believe that Apple could use Gemini’s multi-modal capabilities for its Writings tools , Image Playground , and Message summaries, too.
Elon Musk Not Happy With Apple and Google Tie Up
xAI CEO Elon Musk also responded to Google’s announcement post on X, sharing his thoughts on the matter by saying, “This seems like an unreasonable concentration of power for Google, given that they also have Android and Chrome.” Though we don’t expect either Apple or Google to respond to Elon, we will update the situation as it progresses.
It is worth noting that xAI is the company behind Grok, which is in hot water currently due to its inappropriate image generation fiasco , and has been getting backlash from multiple news outlets, X users, and even government authorities.
With over 4 year of experience under the belt, I cover all facets of consumer tech, from smartphones to other consumer electronics, our favorite social media apps, as well as the growing realm of AI and LLMs. As an Apps and AI writer app Beebom, I provide my expertise in all these areas, weaving stories that help you get familiar with the tech around you. But you will find me playing NYT daily puzzles in my free time.
Add new comment
Name
Email ID
Δ
