RAM Latency Calculator (First Word Latency)

First word latency is how long, in nanoseconds, the memory takes before it starts delivering a new piece of data. The number on the box is a CAS latency and a speed. Those are not the same unit, so a CL30 kit is not automatically slower than a CL16 kit.

Quick Answer: How to Calculate RAM Latency

RAM latency in nanoseconds = CAS latency × 2000 ÷ speed in MT/s. DDR4-3200 CL16 works out to 10 ns, and so does DDR5-6000 CL30, so a higher CL number doesn’t mean slower memory when the speed is higher too. Lower nanoseconds means lower latency. The faster kit at the same latency also gives you more bandwidth, so when two kits tie on nanoseconds, pick the higher speed.

Calculator

Enter the speed in megatransfers per second and the CAS latency from the kit's label. The result is CL × 2000 ÷ MT/s.

Speed (MT/s)

CAS latency

Calculate latency

First word latency: 10.00 ns. This is CL × 2000 ÷ MT/s, not a game benchmark.

The Formula

Nanoseconds: CL × 2000 ÷ MT/s. CAS latency is a count of clock cycles. The speed on the box is the transfer rate. Multiplying the cycle count by 2000 and dividing by the transfer rate converts the count into time. DDR memory transfers twice per clock, which is why the factor is 2000 and not 1000.

Common DDR4 and DDR5 Kits

These are standard advertised XMP and JEDEC labels, converted with the formula. They are not frame-rate results.

Kit	CL	MT/s	First word latency
DDR4-3200 CL16	16	3200	10.00 ns
DDR4-3200 CL14	14	3200	8.75 ns
DDR4-3600 CL18	18	3600	10.00 ns
DDR5-5600 CL40	40	5600	14.29 ns
DDR5-6000 CL30	30	6000	10.00 ns
DDR5-6000 CL28	28	6000	9.33 ns
DDR5-6400 CL32	32	6400	10.00 ns
DDR4-3600 CL16	16	3600	8.89 ns
DDR5-4800 CL40	40	4800	16.67 ns
DDR5-7200 CL34	34	7200	9.44 ns

Lower CL or Higher Speed?

Same latency, different label: compare the nanoseconds, not the CL alone. DDR4-3200 CL16 and DDR5-6000 CL30 are both 10.00 ns. DDR4-3600 CL18 is also 10.00 ns. The CL30 kit looks "looser" and is the same latency as the CL16 kit because the cycles are shorter.

DDR5-5600 CL40 is 14.29 ns. That is a real step slower than DDR5-6000 CL30, even though 5600 is not far behind 6000 on the speed label. This is the case the formula is for.

The RAM bottleneck guide covers capacity, single-channel kits, and speed that is far below what the platform expects. This page only converts latency. The bottleneck calculator uses capacity, speed, type, and channel. It does not ask for CAS latency, so a CL change will not move that result.

Latency vs Bandwidth: Why Speed Still Matters

Latency is only half of memory performance. The other half is bandwidth: how much data the memory moves per second once it starts, which rises with the MT/s number. DDR4-3200 CL16 and DDR5-6000 CL30 have the same 10 ns first word latency, but the DDR5 kit moves data almost twice as fast. That is why a modern DDR5 kit outperforms an old “tight” DDR4 kit in most games even when the nanoseconds are equal.

A simple rule: compare kits by nanoseconds first. If they tie, or are within about half a nanosecond, take the faster MT/s.

Best RAM Speed and CL for Your Platform

• AMD AM4 (Ryzen 3000/5000, DDR4): DDR4-3200 CL16 is the common baseline, and DDR4-3600 CL16 or CL18 is the widely used sweet spot.
• AMD AM5 (Ryzen 7000/9000, DDR5): AMD has described DDR5-6000 as the sweet spot for these chips, and DDR5-6000 CL30 (10 ns) is the most popular kit for them.
• Intel 12th–14th gen and Core Ultra (DDR5): these platforms generally handle higher speeds well, so DDR5-6000 to DDR5-6400 kits at CL30–CL32 are a safe, compatible choice.
• X3D processors: the large on-chip cache serves many requests before they reach system memory, so these chips are generally less sensitive to memory tuning.

How to Check Your RAM Speed and CAS Latency

1. Task Manager: open the Performance tab and select Memory. “Speed” shows the MT/s your RAM is running at right now.
2. CPU-Z: the Memory tab shows the running timings (CAS# Latency), and the SPD tab shows the profiles the kit supports.
3. Compare to the label: if Task Manager shows a lower speed than the box (for example 4800 instead of 6000), XMP or EXPO is off. Turn it on in the BIOS to get the advertised speed and CL.

Running below the rated speed is one of the most common, and easiest to fix, memory bottlenecks. Our RAM bottleneck guide covers it with capacity and single-channel issues.

Does RAM Latency Affect FPS?

It can, in games that are already sensitive to memory, and the size of the change is not something this calculator measures. What it can tell you is whether the kit you are about to buy is actually tighter, or only looks tighter because the CL number is smaller. A kit at the same nanoseconds is not an upgrade in latency.

On AM4, DDR4-3200 CL16 is the usual baseline the rest of the platform was sold with. The best AM4 CPU guide is about which chip to drop in, not which timing to chase. A new AM5 build is in DDR5 territory. Platform fit, including the memory generation the board takes, is part of CPU and GPU compatibility.

Frequently Asked Questions